Basic Filters
2026 |
Andreas Kontogiannis Vasilis Pollatos, Panayotis Mertikopoulos Ioannis Panageas Efficient swap regret minimization in combinatorial bandits Conference Twenty-Ninth Annual Conference on Artificial Intelligence and Statistics (AISTATS 2026), 2026. @conference{Kontogiannis2026, title = {Efficient swap regret minimization in combinatorial bandits}, author = {Andreas Kontogiannis, Vasilis Pollatos, Panayotis Mertikopoulos, Ioannis Panageas}, url = {https://arxiv.org/pdf/2602.02087}, year = {2026}, date = {2026-05-02}, booktitle = {Twenty-Ninth Annual Conference on Artificial Intelligence and Statistics (AISTATS 2026)}, abstract = {This paper addresses the problem of designing efficient no-swap regret algorithms for combinatorial bandits, where the number of actions N is exponentially large in the dimensionality of the problem. In this setting, designing efficient no-swap regret translates to sublinear — in horizon T — swap regret with polylogarithmic dependence on N. In contrast to the weaker notion of external regret minimization – a problem which is fairly well understood in the literature – achieving no-swap regret with a polylogarithmic dependence on N has remained elusive in combinatorial bandits. Our paper resolves this challenge, by introducing a no-swap-regret learning algorithm with regret that scales polylogarithmically in N and is tight for the class of combinatorial bandits. To ground our results, we also demonstrate how to implement the proposed algorithm efficiently — that is, with a per-iteration complexity that also scales polylogarithmically in N — across a wide range of well-studied applications.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } This paper addresses the problem of designing efficient no-swap regret algorithms for combinatorial bandits, where the number of actions N is exponentially large in the dimensionality of the problem. In this setting, designing efficient no-swap regret translates to sublinear — in horizon T — swap regret with polylogarithmic dependence on N. In contrast to the weaker notion of external regret minimization – a problem which is fairly well understood in the literature – achieving no-swap regret with a polylogarithmic dependence on N has remained elusive in combinatorial bandits. Our paper resolves this challenge, by introducing a no-swap-regret learning algorithm with regret that scales polylogarithmically in N and is tight for the class of combinatorial bandits. To ground our results, we also demonstrate how to implement the proposed algorithm efficiently — that is, with a per-iteration complexity that also scales polylogarithmically in N — across a wide range of well-studied applications. |
2025 |
Andreas Soularidis Dimitrios Doumanas, Konstantinos Kotis George Vouros A Automating agentic collaborative ontology engineering with role-playing simulation of LLM-powered agents and RAG technology Journal Article The Knowledge Engineering Review, 40 , pp. e10, 2025. @article{Soularidis2025, title = {Automating agentic collaborative ontology engineering with role-playing simulation of LLM-powered agents and RAG technology}, author = {Andreas Soularidis, Dimitrios Doumanas, Konstantinos Kotis, George A Vouros}, doi = {https://doi.org/10.1017/S026988892510009X}, year = {2025}, date = {2025-12-19}, journal = {The Knowledge Engineering Review}, volume = {40}, pages = {e10}, abstract = {Motivated by the astonishing capabilities of large language models (LLMs) in text-generation, reasoning, and simulation of complex human behaviors, in this paper, we propose a novel multi-component LLM-based framework, namely LLM4ACOE, that fully automates the collaborative ontology engineering (COE) process using role-playing simulation of LLM agents and retrieval augmented generation (RAG) technology. The proposed solution enhances the LLM-powered role-playing simulation with RAG ‘feeding’ the LLM with three different types of external knowledge. This knowledge corresponds to the knowledge required by each of the COE roles (agents), using a component-based framework, as follows: (a) domain-specific data-centric documents, (b) OWL documentation, and (c) ReAct guidelines. The aforementioned components are evaluated in combination, with the aim of investigating their impact on the quality of generated ontologies. The aim of this work is twofold, (a) to identify the capacity of LLM-based agents to generate acceptable (by human-experts) ontologies through agentic collaborative ontology engineering (ACOE) role-playing simulation, at specific levels of acceptance (accuracy, validity, and expressiveness of ontologies) without human intervention and (b) to investigate whether and/or to what extent the selected RAG components affect the quality of the generated ontologies. The evaluation of this novel approach is performed using ChatGPT-o in the domain of search and rescue (SAR) missions. To assess the generated ontologies, quantitative and qualitative measures are employed, focusing on coverage, expressiveness, structure, and human involvement.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Motivated by the astonishing capabilities of large language models (LLMs) in text-generation, reasoning, and simulation of complex human behaviors, in this paper, we propose a novel multi-component LLM-based framework, namely LLM4ACOE, that fully automates the collaborative ontology engineering (COE) process using role-playing simulation of LLM agents and retrieval augmented generation (RAG) technology. The proposed solution enhances the LLM-powered role-playing simulation with RAG ‘feeding’ the LLM with three different types of external knowledge. This knowledge corresponds to the knowledge required by each of the COE roles (agents), using a component-based framework, as follows: (a) domain-specific data-centric documents, (b) OWL documentation, and (c) ReAct guidelines. The aforementioned components are evaluated in combination, with the aim of investigating their impact on the quality of generated ontologies. The aim of this work is twofold, (a) to identify the capacity of LLM-based agents to generate acceptable (by human-experts) ontologies through agentic collaborative ontology engineering (ACOE) role-playing simulation, at specific levels of acceptance (accuracy, validity, and expressiveness of ontologies) without human intervention and (b) to investigate whether and/or to what extent the selected RAG components affect the quality of the generated ontologies. The evaluation of this novel approach is performed using ChatGPT-o in the domain of search and rescue (SAR) missions. To assess the generated ontologies, quantitative and qualitative measures are employed, focusing on coverage, expressiveness, structure, and human involvement. |
Apostolos Glenis, George Vouros Scalable Univariate and Multivariate Time-Series Classifiers with Deep Learning Methods Exploiting Symbolic Representations Journal Article Computers, 14 (12), pp. 563, 2025. @article{Glenis2025, title = {Scalable Univariate and Multivariate Time-Series Classifiers with Deep Learning Methods Exploiting Symbolic Representations}, author = {Apostolos Glenis, George Vouros}, url = {https://www.mdpi.com/2073-431X/14/12/563}, doi = {https://doi.org/10.3390/computers14120563}, year = {2025}, date = {2025-12-17}, journal = {Computers}, volume = {14}, number = {12}, pages = {563}, abstract = {Time-series classification (TSC) is an important task across sciences. Symbolic representations (especially SFA) are very effective at combating noise. In this paper, we employ symbolic representations to create state-of-the-art time-series classifiers, with the aim to advance scalability without sacrificing accuracy. First, we create a graph representation of the time series based on SFA words. We use this representation together with graph kernels and an SVM classifier to create a scalable time-series classifier. Next, we use the graph representation together with a Graph Convolutional Neural Network to test how it fares against state-of-the-art time-series classifiers. Additionally, we devised deep neural networks exploiting the SFA representation, inspired by the text classification domain, to study how they fare against state-of-the-art classifiers. The proposed deep learning classifiers have been adapted and evaluated for the multivariate time-series case and also against state-of-the-art time-series classification algorithms based on symbolic representations.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Time-series classification (TSC) is an important task across sciences. Symbolic representations (especially SFA) are very effective at combating noise. In this paper, we employ symbolic representations to create state-of-the-art time-series classifiers, with the aim to advance scalability without sacrificing accuracy. First, we create a graph representation of the time series based on SFA words. We use this representation together with graph kernels and an SVM classifier to create a scalable time-series classifier. Next, we use the graph representation together with a Graph Convolutional Neural Network to test how it fares against state-of-the-art time-series classifiers. Additionally, we devised deep neural networks exploiting the SFA representation, inspired by the text classification domain, to study how they fare against state-of-the-art classifiers. The proposed deep learning classifiers have been adapted and evaluated for the multivariate time-series case and also against state-of-the-art time-series classification algorithms based on symbolic representations. |
Georgios Bouchouras Dimitrios Doumanas, Andreas Soularidis Konstantinos Kotis George Vouros A Leveraging LLMs for Collaborative Ontology Engineering in Parkinson Disease Monitoring and Alerting Journal Article arXiv, 2025. @article{Bouchouras2025, title = {Leveraging LLMs for Collaborative Ontology Engineering in Parkinson Disease Monitoring and Alerting}, author = {Georgios Bouchouras, Dimitrios Doumanas, Andreas Soularidis, Konstantinos Kotis, George A Vouros}, url = {https://arxiv.org/pdf/2512.14288}, doi = {https://doi.org/10.48550/arXiv.2512.14288}, year = {2025}, date = {2025-12-16}, journal = {arXiv}, abstract = {This paper explores the integration of Large Language Models (LLMs) in the engineering of a Parkinson’s Disease (PD) monitoring and alerting ontology through four key methodologies: One Shot (OS) prompt techniques, Chain of Thought (CoT) prompts, X-HCOME, and SimX-HCOME+. The primary objective is to determine whether LLMs alone can create comprehensive ontologies and, if not, whether human-LLM collaboration can achieve this goal. Consequently, the paper assesses the effectiveness of LLMs in automated ontology development and the enhancement achieved through human-LLM collaboration. Initial ontology generation was performed using One Shot (OS) and Chain of Thought (CoT) prompts, demonstrating the capability of LLMs to autonomously construct ontologies for PD monitoring and alerting. However, these outputs were not comprehensive and required substantial human refinement to enhance their completeness and accuracy. X-HCOME, a hybrid ontology engineering approach that combines human expertise with LLM capabilities, showed significant improvements in ontology comprehensiveness. This methodology resulted in ontologies that are very similar to those constructed by experts. Further experimentation with SimX-HCOME+, another hybrid methodology emphasizing continuous human supervision and iterative refinement, highlighted the importance of ongoing human involvement. This approach led to the creation of more comprehensive and accurate ontologies. Overall, the paper underscores the potential of human-LLM collaboration in advancing ontology engineering, particularly in complex domains like PD. The results suggest promising directions for future research, including the development of specialized GPT models for ontology construction.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This paper explores the integration of Large Language Models (LLMs) in the engineering of a Parkinson’s Disease (PD) monitoring and alerting ontology through four key methodologies: One Shot (OS) prompt techniques, Chain of Thought (CoT) prompts, X-HCOME, and SimX-HCOME+. The primary objective is to determine whether LLMs alone can create comprehensive ontologies and, if not, whether human-LLM collaboration can achieve this goal. Consequently, the paper assesses the effectiveness of LLMs in automated ontology development and the enhancement achieved through human-LLM collaboration. Initial ontology generation was performed using One Shot (OS) and Chain of Thought (CoT) prompts, demonstrating the capability of LLMs to autonomously construct ontologies for PD monitoring and alerting. However, these outputs were not comprehensive and required substantial human refinement to enhance their completeness and accuracy. X-HCOME, a hybrid ontology engineering approach that combines human expertise with LLM capabilities, showed significant improvements in ontology comprehensiveness. This methodology resulted in ontologies that are very similar to those constructed by experts. Further experimentation with SimX-HCOME+, another hybrid methodology emphasizing continuous human supervision and iterative refinement, highlighted the importance of ongoing human involvement. This approach led to the creation of more comprehensive and accurate ontologies. Overall, the paper underscores the potential of human-LLM collaboration in advancing ontology engineering, particularly in complex domains like PD. The results suggest promising directions for future research, including the development of specialized GPT models for ontology construction. |
Andreas Sideras Konstantinos Bougiatiotis, Elias Zavitsanos Georgios Paliouras George Vouros A Multimodal Alignment-Based Anomaly Detection Method for Bankruptcy Prediction Conference Proceedings of the 6th ACM International Conference on AI in Finance, 2025, ISBN: 9798400722202. @conference{Sideras2025, title = {A Multimodal Alignment-Based Anomaly Detection Method for Bankruptcy Prediction}, author = {Andreas Sideras, Konstantinos Bougiatiotis, Elias Zavitsanos, Georgios Paliouras, George Vouros}, url = {https://dl.acm.org/doi/full/10.1145/3768292.3770380}, doi = {https://doi.org/10.1145/3768292.3770380}, isbn = {9798400722202}, year = {2025}, date = {2025-11-15}, booktitle = {Proceedings of the 6th ACM International Conference on AI in Finance}, pages = {53-61}, abstract = {We present a novel anomaly detection method for next-year bankruptcy prediction, utilizing a combination of financial figures and textual content from annual reports. Our approach, MABAD, learns a shared representation space where non-bankrupt firms share position and orientation. Samples that deviate from this pattern are assigned a higher anomaly score. The proposed method is tailored for highly imbalanced scenarios and is robust to heterogeneous, incomplete, and potentially contradictory inputs. We demonstrate that MABAD consistently outperforms a range of strong baselines, and we also curate and release a new publicly available multisource dataset to foster further research in the domain.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } We present a novel anomaly detection method for next-year bankruptcy prediction, utilizing a combination of financial figures and textual content from annual reports. Our approach, MABAD, learns a shared representation space where non-bankrupt firms share position and orientation. Samples that deviate from this pattern are assigned a higher anomaly score. The proposed method is tailored for highly imbalanced scenarios and is robust to heterogeneous, incomplete, and potentially contradictory inputs. We demonstrate that MABAD consistently outperforms a range of strong baselines, and we also curate and release a new publicly available multisource dataset to foster further research in the domain. |
Elias Zavitsanos Konstantinos Bougiatiotis, Andreas Sideras Georgios Paliouras Positive-Unlabeled Learning for Financial Misstatement Detection under Realistic Constraints Conference ICAIF ’25: Proceedings of the 6th ACM International Conference on AI in Finance, 2025, ISBN: 9798400722202. @conference{Zavitsanos2025, title = {Positive-Unlabeled Learning for Financial Misstatement Detection under Realistic Constraints}, author = {Elias Zavitsanos, Konstantinos Bougiatiotis, Andreas Sideras, Georgios Paliouras}, url = {https://dl.acm.org/doi/full/10.1145/3768292.3770366 https://dl.acm.org/doi/epdf/10.1145/3768292.3770366}, doi = {https://doi.org/10.1145/3768292.3770366}, isbn = {9798400722202}, year = {2025}, date = {2025-11-15}, booktitle = {ICAIF ’25: Proceedings of the 6th ACM International Conference on AI in Finance}, pages = {864-872}, abstract = {Detecting financial misstatements is critical for market integrity but remains challenging due to class imbalance, delayed discovery, and limited labeled data. We propose a novel Positive-Unlabeled (PU) learning framework that models the detection task under realistic constraints, where only a small subset of misstatements is known at training time. Our approach integrates unlabeled data into training, preserves temporal structure, and accounts for extreme imbalance. We construct and release a benchmark dataset reflecting these characteristics and evaluate several PU learning methods against recent baselines. Results show that PU-based models consistently outperform supervised approaches, highlighting their suitability for real-world misstatement detection.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } Detecting financial misstatements is critical for market integrity but remains challenging due to class imbalance, delayed discovery, and limited labeled data. We propose a novel Positive-Unlabeled (PU) learning framework that models the detection task under realistic constraints, where only a small subset of misstatements is known at training time. Our approach integrates unlabeled data into training, preserves temporal structure, and accounts for extreme imbalance. We construct and release a benchmark dataset reflecting these characteristics and evaluate several PU learning methods against recent baselines. Results show that PU-based models consistently outperform supervised approaches, highlighting their suitability for real-world misstatement detection. |
Theodore Tranos Nikolaos Fesakis, Thomas Vasileiou Sotirios Christopoulos Georgio Loukos Maria Koutsoupidou 2025 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe), IEEE, 2025, ISBN: 979-8-3315-2503-3. @conference{Tranos2025, title = {AI-Based Energy Forecasting at Different Distribution Grid Levels to Support Baseline Definition and DSO Participation in LFMs}, author = {Theodore Tranos, Nikolaos Fesakis, Thomas Vasileiou, Sotirios Christopoulos, Georgio Loukos, Maria Koutsoupidou}, url = {https://ieeexplore.ieee.org/abstract/document/11305676}, doi = {https://doi.org/10.1109/ISGTEurope64741.2025.11305676}, isbn = {979-8-3315-2503-3}, year = {2025}, date = {2025-10-20}, booktitle = {2025 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe)}, pages = {1-5}, publisher = {IEEE}, abstract = {A crucial aspect of Local Flexibility Markets (LFMs) is the definition of a baseline for energy production and demand forecasting, which serves as a reference for validating and compensating flexibility services. In this study, we explore the application of machine learning techniques, specifically Long Short-Term Memory (LSTM) networks, to establish accurate baselines for consumers and producers connected to the LV grid. The LSTM models leverage real historical demand and generation data from DSO smart meters in Mesogeia, Greece, combined with weather variables such as temperature and cloud coverage, to enhance forecasting accuracy. Our goal is to evaluate forecasting accuracy at the individual participant level and compare it with the accuracy obtained from forecasting on aggregated consumption/production data within a specific grid segment or using data from the secondary substation to which the participants are connected.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } A crucial aspect of Local Flexibility Markets (LFMs) is the definition of a baseline for energy production and demand forecasting, which serves as a reference for validating and compensating flexibility services. In this study, we explore the application of machine learning techniques, specifically Long Short-Term Memory (LSTM) networks, to establish accurate baselines for consumers and producers connected to the LV grid. The LSTM models leverage real historical demand and generation data from DSO smart meters in Mesogeia, Greece, combined with weather variables such as temperature and cloud coverage, to enhance forecasting accuracy. Our goal is to evaluate forecasting accuracy at the individual participant level and compare it with the accuracy obtained from forecasting on aggregated consumption/production data within a specific grid segment or using data from the secondary substation to which the participants are connected. |
Andreas Kontogiannis Vasilis Pollatos, Gabriele Farina Panayotis Mertikopoulos Ioannis Panageas The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025 poster), 2025. @conference{Kontogiannis2025b, title = {Efficient kernelized learning in polyhedral games beyond full-information: From Colonel Blotto to congestion games}, author = {Andreas Kontogiannis, Vasilis Pollatos, Gabriele Farina, Panayotis Mertikopoulos, Ioannis Panageas}, url = {https://openreview.net/attachment?id=FUBaZDMOFj&name=pdf https://arxiv.org/pdf/2509.20919}, doi = {https://doi.org/10.48550/arXiv.2509.20919}, year = {2025}, date = {2025-09-25}, booktitle = {The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025 poster)}, journal = {The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025 poster)}, abstract = {We examine the problem of efficiently learning coarse correlated equilibria (CCE) in polyhedral games, that is, normal-form games with an exponentially large number of actions per player and an underlying combinatorial structure—such as the classic Colonel Blotto game or congestion games. Achieving computational efficiency in this setting requires learning algorithms whose regret and per-iteration complexity scale at most polylogarithmically with the size of the players’ action sets. This challenge has recently been addressed in the full-information setting, primarily through the use of kernelization; however, in the more realistic partial information setting, the situation is much more challenging, and existing approaches result in suboptimal and impractical runtime complexity to learn CCE. We address this gap via a novel kernelization-based framework for payoff-based learning in polyhedral games, which we then apply to certain key classes of polyhedral games—namely Colonel Blotto, graphic matroid and network congestion games. In so doing, we obtain a range of computationally efficient payoff-based learning algorithms which significantly improve upon prior work in terms of the runtime for learning CCE.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } We examine the problem of efficiently learning coarse correlated equilibria (CCE) in polyhedral games, that is, normal-form games with an exponentially large number of actions per player and an underlying combinatorial structure—such as the classic Colonel Blotto game or congestion games. Achieving computational efficiency in this setting requires learning algorithms whose regret and per-iteration complexity scale at most polylogarithmically with the size of the players’ action sets. This challenge has recently been addressed in the full-information setting, primarily through the use of kernelization; however, in the more realistic partial information setting, the situation is much more challenging, and existing approaches result in suboptimal and impractical runtime complexity to learn CCE. We address this gap via a novel kernelization-based framework for payoff-based learning in polyhedral games, which we then apply to certain key classes of polyhedral games—namely Colonel Blotto, graphic matroid and network congestion games. In so doing, we obtain a range of computationally efficient payoff-based learning algorithms which significantly improve upon prior work in terms of the runtime for learning CCE. |
Theocharis Kravaris, George Vouros A Transferable aircraft trajectory prediction with generative deep imitation learning Journal Article International Journal of Data Science and Analytics, 20 (3), pp. 1977-1999, 2025. @article{Kravaris2025, title = {Transferable aircraft trajectory prediction with generative deep imitation learning}, author = {Theocharis Kravaris, George A Vouros}, url = {https://link.springer.com/article/10.1007/s41060-024-00574-1}, doi = {https://doi.org/10.1007/s41060-024-00574-1}, year = {2025}, date = {2025-09-01}, journal = {International Journal of Data Science and Analytics}, volume = {20}, number = {3}, pages = {1977-1999}, abstract = {Trajectory-oriented transformations to air traffic management operations require high fidelity aircraft trajectory prediction capabilities. Data-driven trajectory prediction approaches provide promising results, albeit with important limitations that hinder seriously the efficient and effective deployment of trajectory prediction methods: They need abundant training effort with a large amount of training samples and require training distinct models for different origin–destination (OD) airport pairs. In this paper, we address the problem of building transferable trajectory prediction models, casting the prediction problem as a transferable imitation task, introducing a novel formulation which (a) provides the capability to utilize trained models, in new OD pairs, offering a warm starting for computationally efficient training, and (b) improves the efficacy of data-driven trajectory prediction. The proposed approach provides very accurate results for large look-ahead time predictions, even if transferable models have been trained with few samples.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Trajectory-oriented transformations to air traffic management operations require high fidelity aircraft trajectory prediction capabilities. Data-driven trajectory prediction approaches provide promising results, albeit with important limitations that hinder seriously the efficient and effective deployment of trajectory prediction methods: They need abundant training effort with a large amount of training samples and require training distinct models for different origin–destination (OD) airport pairs. In this paper, we address the problem of building transferable trajectory prediction models, casting the prediction problem as a transferable imitation task, introducing a novel formulation which (a) provides the capability to utilize trained models, in new OD pairs, offering a warm starting for computationally efficient training, and (b) improves the efficacy of data-driven trajectory prediction. The proposed approach provides very accurate results for large look-ahead time predictions, even if transferable models have been trained with few samples. |
George Papadopoulos, George Vouros A Learning safe, constrained policies via imitation learning: Connection to Probabilistic Inference and a Naive Algorithm Journal Article arXiv, 2025. @article{Papadopoulos2025b, title = {Learning safe, constrained policies via imitation learning: Connection to Probabilistic Inference and a Naive Algorithm}, author = {George Papadopoulos, George A Vouros}, url = {https://arxiv.org/pdf/2507.06780}, doi = {https://doi.org/10.48550/arXiv.2507.06780}, year = {2025}, date = {2025-07-09}, journal = {arXiv}, abstract = {This article introduces an imitation learning method for learning maximum entropy policies that comply with constraints demonstrated by expert trajectories executing a task. The formulation of the method takes advantage of results connecting performance to bounds for the KL-divergence between demonstrated and learned policies, and its objective is rigorously justified through a connection to a probabilistic inference framework for reinforcement learning, incorporating the reinforcement learning objective and the objective to abide by constraints in an entropy maximization setting. The proposed algorithm optimizes the learning objective with dual gradient descent, supporting effective and stable training. Experiments show that the proposed method can learn effective policy models for constraints-abiding behaviour, in settings with multiple constraints of different types, accommodating different modalities of demonstrated behaviour, and with abilities to generalize.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This article introduces an imitation learning method for learning maximum entropy policies that comply with constraints demonstrated by expert trajectories executing a task. The formulation of the method takes advantage of results connecting performance to bounds for the KL-divergence between demonstrated and learned policies, and its objective is rigorously justified through a connection to a probabilistic inference framework for reinforcement learning, incorporating the reinforcement learning objective and the objective to abide by constraints in an entropy maximization setting. The proposed algorithm optimizes the learning objective with dual gradient descent, supporting effective and stable training. Experiments show that the proposed method can learn effective policy models for constraints-abiding behaviour, in settings with multiple constraints of different types, accommodating different modalities of demonstrated behaviour, and with abilities to generalize. |
Piyabhum Chaysri Theodoros Tranos, George Papadopoulos George Vouros Konstantinos Blekas A Efficient Autonomous Marine Vessel Navigation with Safe Deep Reinforcement Learning Conference 2025 Symposium on Maritime Informatics and Robotics (MARIS), 2025. @conference{Chaysri2025, title = { Efficient Autonomous Marine Vessel Navigation with Safe Deep Reinforcement Learning}, author = {Piyabhum Chaysri, Theodoros Tranos, George Papadopoulos, George A Vouros, Konstantinos Blekas}, doi = {https://doi.org/10.1109/MARIS64137.2025.11139786}, year = {2025}, date = {2025-06-26}, booktitle = {2025 Symposium on Maritime Informatics and Robotics (MARIS)}, abstract = {The rise of automation and self-driving systems brings a strong focus on safety-centric decision-making, especially in complex environments with large degree of uncertainty where unpredictable interactions occur at high frequency. In this study we address the challenge of safe and efficient maritime navigation by proposing a safe Deep Reinforcement Learning scheme for training Unmanned Surface Vehicle (USV) agents. Our approach leverages the Lagrangian relaxation framework to effectively handle safety constraints, ensuring that the learned navigation policies balance goal achievement with obstacle avoidance. We address each type of static and moving obstacle separately with the aim of achieving more effective management of their impact on safe navigation. This enables the design of a more advanced constraint-aware optimization framework, enhancing USV’s ability to navigate complex maritime environment, adapt to changing traffic conditions and maintain a minimal risk of collision. Experiments were conducted in a simulated environment tailored to match realistic weather and traffic density conditions. The simulation results highlight the potential of the proposed method in developing advanced USV navigation policies that achieve high accuracy and enhanced safety.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } The rise of automation and self-driving systems brings a strong focus on safety-centric decision-making, especially in complex environments with large degree of uncertainty where unpredictable interactions occur at high frequency. In this study we address the challenge of safe and efficient maritime navigation by proposing a safe Deep Reinforcement Learning scheme for training Unmanned Surface Vehicle (USV) agents. Our approach leverages the Lagrangian relaxation framework to effectively handle safety constraints, ensuring that the learned navigation policies balance goal achievement with obstacle avoidance. We address each type of static and moving obstacle separately with the aim of achieving more effective management of their impact on safe navigation. This enables the design of a more advanced constraint-aware optimization framework, enhancing USV’s ability to navigate complex maritime environment, adapt to changing traffic conditions and maintain a minimal risk of collision. Experiments were conducted in a simulated environment tailored to match realistic weather and traffic density conditions. The simulation results highlight the potential of the proposed method in developing advanced USV navigation policies that achieve high accuracy and enhanced safety. |
George Giannakopoulos Andreas Sideras, Konstantinos Stamatakis Nikolaos Melanitis NAVMAT: An AI-supported naval failures knowledge management system Journal Article Expert Systems with Applications, 277 , pp. 127117, 2025. @article{Giannakopoulos2025, title = {NAVMAT: An AI-supported naval failures knowledge management system}, author = {George Giannakopoulos, Andreas Sideras, Konstantinos Stamatakis, Nikolaos Melanitis}, doi = {https://doi.org/10.1016/j.eswa.2025.127117}, year = {2025}, date = {2025-06-05}, journal = {Expert Systems with Applications}, volume = {277}, pages = {127117}, abstract = {We present “NAVMAT”, an intelligent, multilingual knowledge management platform designed to record and categorize material failure incidents reported in naval operations. This paper provides an overview of the platform, identifying its key software components and highlighting the information retrieval approach used to support user workflows. The platform primarily facilitates real-time, multilingual search and intelligent indexing, streamlining the incident management process while offering valuable insights from past incidents and knowledge resources. To achieve this, it employs a customized natural language processing pipeline integrated with a carefully engineered ontology. The ontology, regularly updated by domain experts, enriches the retrieval mechanism by instilling domain specific knowledge. This approach aims to reduce the significant variability in specialized terminology by promoting convergence towards a unified vocabulary.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We present “NAVMAT”, an intelligent, multilingual knowledge management platform designed to record and categorize material failure incidents reported in naval operations. This paper provides an overview of the platform, identifying its key software components and highlighting the information retrieval approach used to support user workflows. The platform primarily facilitates real-time, multilingual search and intelligent indexing, streamlining the incident management process while offering valuable insights from past incidents and knowledge resources. To achieve this, it employs a customized natural language processing pipeline integrated with a carefully engineered ontology. The ontology, regularly updated by domain experts, enriches the retrieval mechanism by instilling domain specific knowledge. This approach aims to reduce the significant variability in specialized terminology by promoting convergence towards a unified vocabulary. |
Andreas Kontogiannis Konstantinos Papathanasiou, Yi Shen Giorgos Stamou Michael Zavlanos George Vouros M Enhancing cooperative multi-agent reinforcement learning with state modelling and adversarial exploration Journal Article arXiv, 2025. @article{Kontogiannis2025, title = {Enhancing cooperative multi-agent reinforcement learning with state modelling and adversarial exploration}, author = {Andreas Kontogiannis, Konstantinos Papathanasiou, Yi Shen, Giorgos Stamou, Michael M Zavlanos, George Vouros}, url = {https://arxiv.org/pdf/2505.05262}, doi = {https://doi.org/10.48550/arXiv.2505.05262}, year = {2025}, date = {2025-05-08}, journal = {arXiv}, abstract = {Learning to cooperate in distributed partially observable environments with no communication abilities poses significant challenges for multi-agent deep reinforcement learning (MARL). This paper addresses key concerns in this domain, focusing on inferring state representations from individual agent observations and leveraging these representations to enhance agents’ exploration and collaborative task execution policies. To this end, we propose a novel state modelling framework for cooperative MARL, where agents infer meaningful belief representations of the non-observable state, with respect to optimizing their own policies, while filtering redundant and less informative joint state information. Building upon this framework, we propose the MARL SMPE algorithm. In SMPE, agents enhance their own policy’s discriminative abilities under partial observability, explicitly by incorporating their beliefs into the policy network, and implicitly by adopting an adversarial type of exploration policies which encourages agents to discover novel, high-value states while improving the discriminative abilities of others. Experimentally, we show that SMPE outperforms state-of-the-art MARL algorithms in complex fully cooperative tasks from the MPE, LBF, and RWARE benchmarks.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Learning to cooperate in distributed partially observable environments with no communication abilities poses significant challenges for multi-agent deep reinforcement learning (MARL). This paper addresses key concerns in this domain, focusing on inferring state representations from individual agent observations and leveraging these representations to enhance agents’ exploration and collaborative task execution policies. To this end, we propose a novel state modelling framework for cooperative MARL, where agents infer meaningful belief representations of the non-observable state, with respect to optimizing their own policies, while filtering redundant and less informative joint state information. Building upon this framework, we propose the MARL SMPE algorithm. In SMPE, agents enhance their own policy’s discriminative abilities under partial observability, explicitly by incorporating their beliefs into the policy network, and implicitly by adopting an adversarial type of exploration policies which encourages agents to discover novel, high-value states while improving the discriminative abilities of others. Experimentally, we show that SMPE outperforms state-of-the-art MARL algorithms in complex fully cooperative tasks from the MPE, LBF, and RWARE benchmarks. |
Despoina P Kiouri Georgios C Batsis, Thomas Mavromoustakos Alessandro Giuliani Christos Chasapis T Structure-Based Modeling of the Gut Bacteria–Host Interactome Through Statistical Analysis of Domain–Domain Associations Using Machine Learning Journal Article BioTech, 14 (1), pp. 13, 2025. @article{Kiouri2025c, title = {Structure-Based Modeling of the Gut Bacteria–Host Interactome Through Statistical Analysis of Domain–Domain Associations Using Machine Learning}, author = {Despoina P Kiouri, Georgios C Batsis, Thomas Mavromoustakos, Alessandro Giuliani, Christos T Chasapis}, url = {https://www.mdpi.com/2673-6284/14/1/13/pdf?version=1740479495}, doi = {https://doi.org/10.3390/biotech14010013}, year = {2025}, date = {2025-02-25}, journal = {BioTech}, volume = {14}, number = {1}, pages = {13}, abstract = {The gut microbiome, a complex ecosystem of microorganisms, plays a pivotal role in human health and disease. The gut microbiome’s influence extends beyond the digestive system to various organs, and its imbalance is linked to a wide range of diseases, including cancer and neurodevelopmental, inflammatory, metabolic, cardiovascular, autoimmune, and psychiatric diseases. Despite its significance, the interactions between gut bacteria and human proteins remain understudied, with less than 20,000 experimentally validated protein interactions between the host and any bacteria species. This study addresses this knowledge gap by predicting a protein–protein interaction network between gut bacterial and human proteins. Using statistical associations between Pfam domains, a comprehensive dataset of over one million experimentally validated pan-bacterial–human protein interactions, as well as inter- and intra-species protein interactions from various organisms, were used for the development of a machine learning-based prediction method to uncover key regulatory molecules in this dynamic system. This study’s findings contribute to the understanding of the intricate gut microbiome–host relationship and pave the way for future experimental validation and therapeutic strategies targeting the gut microbiome interplay.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The gut microbiome, a complex ecosystem of microorganisms, plays a pivotal role in human health and disease. The gut microbiome’s influence extends beyond the digestive system to various organs, and its imbalance is linked to a wide range of diseases, including cancer and neurodevelopmental, inflammatory, metabolic, cardiovascular, autoimmune, and psychiatric diseases. Despite its significance, the interactions between gut bacteria and human proteins remain understudied, with less than 20,000 experimentally validated protein interactions between the host and any bacteria species. This study addresses this knowledge gap by predicting a protein–protein interaction network between gut bacterial and human proteins. Using statistical associations between Pfam domains, a comprehensive dataset of over one million experimentally validated pan-bacterial–human protein interactions, as well as inter- and intra-species protein interactions from various organisms, were used for the development of a machine learning-based prediction method to uncover key regulatory molecules in this dynamic system. This study’s findings contribute to the understanding of the intricate gut microbiome–host relationship and pave the way for future experimental validation and therapeutic strategies targeting the gut microbiome interplay. |
Natalia Koliou, George Vouros Ranking Joint Policies in Dynamic Games using Evolutionary Dynamics Journal Article arXiv, 2025. @article{Koliou2025, title = {Ranking Joint Policies in Dynamic Games using Evolutionary Dynamics}, author = {Natalia Koliou, George Vouros}, url = {https://arxiv.org/pdf/2502.14724}, doi = {https://doi.org/10.48550/arXiv.2502.14724}, year = {2025}, date = {2025-02-20}, journal = {arXiv}, abstract = {Game-theoretic solution concepts, such as the Nash equilibrium, have been key to finding stable joint actions in multi-player games. However, it has been shown that the dynamics of agents’ interactions, even in simple two-player games with few strategies, are incapable of reaching Nash equilibria, exhibiting complex and unpredictable behavior. Instead, evolutionary approaches can describe the long-term persistence of strategies and filter out transient ones, accounting for the long-term dynamics of agents’ interactions. Our goal is to identify agents’ joint strategies that result in stable behavior, being resistant to changes, while also accounting for agents’ payoffs, in dynamic games. Towards this goal, and building on previous results, this paper proposes transforming dynamic games into their empirical forms by considering agents’ strategies instead of agents’ actions, and applying the evolutionary methodology -Rank to evaluate and rank strategy profiles according to their long-term dynamics. This methodology not only allows us to identify joint strategies that are strong through agents’ long-term interactions, but also provides a descriptive, transparent framework regarding the high ranking of these strategies. Experiments report on agents that aim to collaboratively solve a stochastic version of the graph coloring problem. We consider different styles of play as strategies to define the empirical game, and train policies realizing these strategies, using the DQN algorithm. Then we run simulations to generate the payoff matrix required by -Rank to rank joint strategies.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Game-theoretic solution concepts, such as the Nash equilibrium, have been key to finding stable joint actions in multi-player games. However, it has been shown that the dynamics of agents’ interactions, even in simple two-player games with few strategies, are incapable of reaching Nash equilibria, exhibiting complex and unpredictable behavior. Instead, evolutionary approaches can describe the long-term persistence of strategies and filter out transient ones, accounting for the long-term dynamics of agents’ interactions. Our goal is to identify agents’ joint strategies that result in stable behavior, being resistant to changes, while also accounting for agents’ payoffs, in dynamic games. Towards this goal, and building on previous results, this paper proposes transforming dynamic games into their empirical forms by considering agents’ strategies instead of agents’ actions, and applying the evolutionary methodology -Rank to evaluate and rank strategy profiles according to their long-term dynamics. This methodology not only allows us to identify joint strategies that are strong through agents’ long-term interactions, but also provides a descriptive, transparent framework regarding the high ranking of these strategies. Experiments report on agents that aim to collaboratively solve a stochastic version of the graph coloring problem. We consider different styles of play as strategies to define the empirical game, and train policies realizing these strategies, using the DQN algorithm. Then we run simulations to generate the payoff matrix required by -Rank to rank joint strategies. |
Christos Spatharis Konstantinos Blekas, George Santipantakis George Vouros Modular and Multimodal Generative Adversarial Imitation Learning for Modeling Flight Trajectories Journal Article Journal of Air Transportation, 33 (3), pp. 188-204, 2025. @article{Spatharis2025, title = {Modular and Multimodal Generative Adversarial Imitation Learning for Modeling Flight Trajectories}, author = {Christos Spatharis, Konstantinos Blekas, George Santipantakis, George Vouros}, doi = {https://doi.org/10.2514/1.D0396}, year = {2025}, date = {2025-02-17}, journal = {Journal of Air Transportation}, volume = {33}, number = {3}, pages = {188-204}, abstract = {We aim to imitate the execution of modular tasks by exploiting unsegmented trajectories that demonstrate the execution of these tasks. This is challenging since the execution of tasks follows different modes (i.e., patterns of behavior), which may exist in various mixtures within subtasks, and the identification of trajectories’ modules (i.e., subtrajectories executing subtasks) may not be easy. This paper addresses the modularity of trajectories in conjunction with multimodality toward imitating the execution of aircraft trajectories. It proposes an imitation learning framework for the aircraft trajectory prediction problem, which segments demonstrated aircraft trajectories into subtrajectories corresponding to flight phases. This facilitates disentangling modes and learning a mixture of policies per flight phase. While trajectories are segmented using domain-specific rules, a mixture of policies per flight phase is learned by a generative multimodal imitation learning method. This modular approach enables accurate prediction of both modes and subtrajectories, which finally results in predicting the evolution of the aircraft state across the whole trajectory in a compositional way. Experiments using a real-world dataset of long flights show the potential of the proposed framework to disentangle multimodal trajectories in real-world settings and predict trajectories with high accuracy, in comparison to methods that do not exploit subtrajectories.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We aim to imitate the execution of modular tasks by exploiting unsegmented trajectories that demonstrate the execution of these tasks. This is challenging since the execution of tasks follows different modes (i.e., patterns of behavior), which may exist in various mixtures within subtasks, and the identification of trajectories’ modules (i.e., subtrajectories executing subtasks) may not be easy. This paper addresses the modularity of trajectories in conjunction with multimodality toward imitating the execution of aircraft trajectories. It proposes an imitation learning framework for the aircraft trajectory prediction problem, which segments demonstrated aircraft trajectories into subtrajectories corresponding to flight phases. This facilitates disentangling modes and learning a mixture of policies per flight phase. While trajectories are segmented using domain-specific rules, a mixture of policies per flight phase is learned by a generative multimodal imitation learning method. This modular approach enables accurate prediction of both modes and subtrajectories, which finally results in predicting the evolution of the aircraft state across the whole trajectory in a compositional way. Experiments using a real-world dataset of long flights show the potential of the proposed framework to disentangle multimodal trajectories in real-world settings and predict trajectories with high accuracy, in comparison to methods that do not exploit subtrajectories. |
Despoina P Kiouri Georgios C Batsis, Christos Chasapis T Structure-Based Deep Learning Framework for Modeling Human–Gut Bacterial Protein Interactions Journal Article Proteomes, 13 (1), pp. 10, 2025. @article{Kiouri2025b, title = {Structure-Based Deep Learning Framework for Modeling Human–Gut Bacterial Protein Interactions}, author = {Despoina P Kiouri, Georgios C Batsis, Christos T Chasapis}, url = {https://www.mdpi.com/2227-7382/13/1/10/pdf?version=1739790880}, doi = {https://doi.org/10.3390/proteomes13010010}, year = {2025}, date = {2025-02-17}, journal = {Proteomes}, volume = {13}, number = {1}, pages = {10}, abstract = {The interaction network between the human host proteins and the proteins of the gut bacteria is essential for the establishment of human health, and its dysregulation directly contributes to disease development. Despite its great importance, experimental data on protein–protein interactions (PPIs) between these species are sparse due to experimental limitations. Methods: This study presents a deep learning-based framework for predicting PPIs between human and gut bacterial proteins using structural data. The framework leverages graph-based protein representations and variational autoencoders (VAEs) to extract structural embeddings from protein graphs, which are then fused through a Bi-directional Cross-Attention module to predict interactions. The model addresses common challenges in PPI datasets, such as class imbalance, using focal loss to emphasize harder-to-classify samples. Results: The results demonstrated that this framework exhibits robust performance, with high precision and recall across validation and test datasets, underscoring its generalizability. By incorporating proteoforms in the analysis, the model accounts for the structural complexity within proteomes, making predictions biologically relevant. Conclusions: These findings offer a scalable tool for investigating the interactions between the host and the gut microbiota, potentially yielding new treatment targets and diagnostics for disorders linked to the microbiome.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The interaction network between the human host proteins and the proteins of the gut bacteria is essential for the establishment of human health, and its dysregulation directly contributes to disease development. Despite its great importance, experimental data on protein–protein interactions (PPIs) between these species are sparse due to experimental limitations. Methods: This study presents a deep learning-based framework for predicting PPIs between human and gut bacterial proteins using structural data. The framework leverages graph-based protein representations and variational autoencoders (VAEs) to extract structural embeddings from protein graphs, which are then fused through a Bi-directional Cross-Attention module to predict interactions. The model addresses common challenges in PPI datasets, such as class imbalance, using focal loss to emphasize harder-to-classify samples. Results: The results demonstrated that this framework exhibits robust performance, with high precision and recall across validation and test datasets, underscoring its generalizability. By incorporating proteoforms in the analysis, the model accounts for the structural complexity within proteomes, making predictions biologically relevant. Conclusions: These findings offer a scalable tool for investigating the interactions between the host and the gut microbiota, potentially yielding new treatment targets and diagnostics for disorders linked to the microbiome. |
George Papadopoulos Andreas Kontogiannis, Foteini Papadopoulou Chaido Poulianou Ioannis Koumentis George Vouros An extended benchmarking of multi-agent reinforcement learning algorithms in complex fully cooperative tasks Journal Article arXiv, 2025. @article{Papadopoulos2025, title = {An extended benchmarking of multi-agent reinforcement learning algorithms in complex fully cooperative tasks}, author = {George Papadopoulos, Andreas Kontogiannis, Foteini Papadopoulou, Chaido Poulianou, Ioannis Koumentis, George Vouros}, url = {https://arxiv.org/pdf/2502.04773}, doi = {https://doi.org/10.48550/arXiv.2502.04773}, year = {2025}, date = {2025-02-07}, journal = {arXiv}, abstract = {Multi-Agent Reinforcement Learning (MARL) has recently emerged as a significant area of research. However, MARL evaluation often lacks systematic diversity, hindering a comprehensive understanding of algorithms’ capabilities. In particular, cooperative MARL algorithms are predominantly evaluated on benchmarks such as SMAC and GRF, which primarily feature team game scenarios without assessing adequately various aspects of agents’ capabilities required in fully cooperative real-world tasks such as multi-robot cooperation and warehouse, resource management, search and rescue, and human-AI cooperation. Moreover, MARL algorithms are mainly evaluated on low dimensional state spaces, and thus their performance on high-dimensional (e.g., image) observations is not well-studied. To fill this gap, this paper highlights the crucial need for expanding systematic evaluation across a wider array of existing benchmarks. To this end, we conduct extensive evaluation and comparisons of well-known MARL algorithms on complex fully cooperative benchmarks, including tasks with images as agents’ observations. Interestingly, our analysis shows that many algorithms, hailed as state-of-the-art on SMAC and GRF, may underperform standard MARL baselines on fully cooperative benchmarks. Finally, towards more systematic and better evaluation of cooperative MARL algorithms, we have open-sourced PyMARLzoo+, an extension of the widely used (E)PyMARL libraries, which addresses an open challenge from [49], facilitating seamless integration and support with all benchmarks of PettingZoo, as well as Overcooked, PressurePlate, Capture Target and Box Pushing.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Multi-Agent Reinforcement Learning (MARL) has recently emerged as a significant area of research. However, MARL evaluation often lacks systematic diversity, hindering a comprehensive understanding of algorithms’ capabilities. In particular, cooperative MARL algorithms are predominantly evaluated on benchmarks such as SMAC and GRF, which primarily feature team game scenarios without assessing adequately various aspects of agents’ capabilities required in fully cooperative real-world tasks such as multi-robot cooperation and warehouse, resource management, search and rescue, and human-AI cooperation. Moreover, MARL algorithms are mainly evaluated on low dimensional state spaces, and thus their performance on high-dimensional (e.g., image) observations is not well-studied. To fill this gap, this paper highlights the crucial need for expanding systematic evaluation across a wider array of existing benchmarks. To this end, we conduct extensive evaluation and comparisons of well-known MARL algorithms on complex fully cooperative benchmarks, including tasks with images as agents’ observations. Interestingly, our analysis shows that many algorithms, hailed as state-of-the-art on SMAC and GRF, may underperform standard MARL baselines on fully cooperative benchmarks. Finally, towards more systematic and better evaluation of cooperative MARL algorithms, we have open-sourced PyMARLzoo+, an extension of the widely used (E)PyMARL libraries, which addresses an open challenge from [49], facilitating seamless integration and support with all benchmarks of PettingZoo, as well as Overcooked, PressurePlate, Capture Target and Box Pushing. |
Fotis Assimakopoulos Costas Vassilakis, Dionisis Margaris Konstantinos Kotis Dimitris Spiliotopoulos AI and related technologies in the fields of smart agriculture: A review Journal Article Information, 16 (2), pp. 100, 2025. @article{Assimakopoulos2025, title = {AI and related technologies in the fields of smart agriculture: A review}, author = {Fotis Assimakopoulos, Costas Vassilakis, Dionisis Margaris, Konstantinos Kotis, Dimitris Spiliotopoulos}, url = {https://www.mdpi.com/2078-2489/16/2/100/pdf?version=1738918054}, doi = {https://doi.org/10.3390/info16020100}, year = {2025}, date = {2025-02-02}, journal = {Information}, volume = {16}, number = {2}, pages = {100}, abstract = {The integration of cutting-edge technologies—such as the Internet of Things (IoT), artificial intelligence (AI), machine learning (ML), and various emerging technologies—is revolutionizing agricultural practices, enhancing productivity, sustainability, and efficiency. The objective of this study is to review the literature regarding the development and evolution of AI as well as other emerging technologies in the various fields of Agriculture as they are developed and transformed by integrating the above technologies. The areas examined in this study are open field smart farming, vertical and indoor farming, zero waste agriculture, precision livestock farming, smart greenhouses, and regenerative agriculture. This paper links current research, technological innovations, and case studies to present a comprehensive review of these emerging technologies being developed in the context of smart agriculture, for the benefit of farmers and consumers in general. By exploring practical applications and future perspectives, this work aims to provide valuable insights to address global food security challenges, minimize environmental impacts, and support sustainable development goals through the application of new technologies.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The integration of cutting-edge technologies—such as the Internet of Things (IoT), artificial intelligence (AI), machine learning (ML), and various emerging technologies—is revolutionizing agricultural practices, enhancing productivity, sustainability, and efficiency. The objective of this study is to review the literature regarding the development and evolution of AI as well as other emerging technologies in the various fields of Agriculture as they are developed and transformed by integrating the above technologies. The areas examined in this study are open field smart farming, vertical and indoor farming, zero waste agriculture, precision livestock farming, smart greenhouses, and regenerative agriculture. This paper links current research, technological innovations, and case studies to present a comprehensive review of these emerging technologies being developed in the context of smart agriculture, for the benefit of farmers and consumers in general. By exploring practical applications and future perspectives, this work aims to provide valuable insights to address global food security challenges, minimize environmental impacts, and support sustainable development goals through the application of new technologies. |
Dimitrios Doumanas Georgios Bouchouras, Andreas Soularidis Konstantinos Kotis George Vouros From human-to LLM-centered collaborative ontology engineering Journal Article Applied Ontology, 19 (4), pp. 334-367, 2025. @article{Doumanas2025, title = {From human-to LLM-centered collaborative ontology engineering}, author = {Dimitrios Doumanas, Georgios Bouchouras, Andreas Soularidis, Konstantinos Kotis, George Vouros}, doi = {https://doi.org/10.1177/15705838241305067}, year = {2025}, date = {2025-01-31}, journal = {Applied Ontology}, volume = {19}, number = {4}, pages = {334-367}, abstract = {In the continuously evolving landscape of knowledge engineering, the symbiosis and teaming of humans and machines emerge as a pivotal new domain. This article explores the multifaceted realms of human and machine collaborative ontology engineering (OE). The goal of the presented work is to explore the potential of Large Language Models (LLMs) to speed up and automate the processes of collaborative OE, experimenting with different levels of LLM involvement. The proposed approach is based on a human-centered approach, that is, the HCOME approach to collaborative OE, and follows a process of exploring the declining involvement of humans and the parallel increase of LLM involvement, concluding at a level of automation where the OE is exclusively performed by LLMs. This experimentation is organized based on a series of human/LLM collaboration levels (a spectrum of OE), each one aligned to a specific OE methodology, that is, Level-0 HCOME (Human), Level-1 X-HCOME (Human and LLMs), Level-2 SimX-HCOME (LLMs and Human), and Level-3 Sim-HCOME (LLMs). The evaluation of these methodologies (one per level) is performed by measuring the similarity of the generated ontologies against “reference” ontologies (precision, recall, and F1-score of reference-to-LLM-generated ontological mappings). The results presented in this paper demonstrate that while LLMs significantly expedite the OE process, the accuracy and completeness of the resulting ontologies are notably enhanced by maintaining a high level of human involvement. This study is expected to contribute to a deeper understanding of evolving dynamics in LLM-based/enhanced OE, paving the way for future advancements toward more effective collaborative OE frameworks.}, keywords = {}, pubstate = {published}, tppubtype = {article} } In the continuously evolving landscape of knowledge engineering, the symbiosis and teaming of humans and machines emerge as a pivotal new domain. This article explores the multifaceted realms of human and machine collaborative ontology engineering (OE). The goal of the presented work is to explore the potential of Large Language Models (LLMs) to speed up and automate the processes of collaborative OE, experimenting with different levels of LLM involvement. The proposed approach is based on a human-centered approach, that is, the HCOME approach to collaborative OE, and follows a process of exploring the declining involvement of humans and the parallel increase of LLM involvement, concluding at a level of automation where the OE is exclusively performed by LLMs. This experimentation is organized based on a series of human/LLM collaboration levels (a spectrum of OE), each one aligned to a specific OE methodology, that is, Level-0 HCOME (Human), Level-1 X-HCOME (Human and LLMs), Level-2 SimX-HCOME (LLMs and Human), and Level-3 Sim-HCOME (LLMs). The evaluation of these methodologies (one per level) is performed by measuring the similarity of the generated ontologies against “reference” ontologies (precision, recall, and F1-score of reference-to-LLM-generated ontological mappings). The results presented in this paper demonstrate that while LLMs significantly expedite the OE process, the accuracy and completeness of the resulting ontologies are notably enhanced by maintaining a high level of human involvement. This study is expected to contribute to a deeper understanding of evolving dynamics in LLM-based/enhanced OE, paving the way for future advancements toward more effective collaborative OE frameworks. |
Despoina P Kiouri Georgios C Batsis, Christos Chasapis T Structure-based approaches for protein–protein interaction prediction using machine learning and deep learning Journal Article Biomolecules, 15 (1), pp. 141, 2025. @article{Kiouri2025, title = {Structure-based approaches for protein–protein interaction prediction using machine learning and deep learning}, author = {Despoina P Kiouri, Georgios C Batsis, Christos T Chasapis}, url = {https://www.mdpi.com/2218-273X/15/1/141/pdf?version=1737097468}, doi = {https://doi.org/10.3390/biom15010141}, year = {2025}, date = {2025-01-17}, journal = {Biomolecules}, volume = {15}, number = {1}, pages = {141}, abstract = {Protein–Protein Interaction (PPI) prediction plays a pivotal role in understanding cellular processes and uncovering molecular mechanisms underlying health and disease. Structure-based PPI prediction has emerged as a robust alternative to sequence-based methods, offering greater biological accuracy by integrating three-dimensional spatial and biochemical features. This work summarizes the recent advances in computational approaches leveraging protein structure information for PPI prediction, focusing on machine learning (ML) and deep learning (DL) techniques. These methods not only improve predictive accuracy but also provide insights into functional sites, such as binding and catalytic residues. However, challenges such as limited high-resolution structural data and the need for effective negative sampling persist. Through the integration of experimental and computational tools, structure-based prediction paves the way for comprehensive proteomic network analysis, holding promise for advancements in drug discovery, biomarker identification, and personalized medicine. Future directions include enhancing scalability and dataset reliability to expand these approaches across diverse proteomes.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Protein–Protein Interaction (PPI) prediction plays a pivotal role in understanding cellular processes and uncovering molecular mechanisms underlying health and disease. Structure-based PPI prediction has emerged as a robust alternative to sequence-based methods, offering greater biological accuracy by integrating three-dimensional spatial and biochemical features. This work summarizes the recent advances in computational approaches leveraging protein structure information for PPI prediction, focusing on machine learning (ML) and deep learning (DL) techniques. These methods not only improve predictive accuracy but also provide insights into functional sites, such as binding and catalytic residues. However, challenges such as limited high-resolution structural data and the need for effective negative sampling persist. Through the integration of experimental and computational tools, structure-based prediction paves the way for comprehensive proteomic network analysis, holding promise for advancements in drug discovery, biomarker identification, and personalized medicine. Future directions include enhancing scalability and dataset reliability to expand these approaches across diverse proteomes. |
Georgios Bouchouras, Konstantinos Kotis Algorithms, 18 (1), pp. 34, 2025. @article{Bouchouras2025b, title = {Integrating artificial intelligence, internet of things, and sensor-based technologies: a systematic review of methodologies in autism Spectrum disorder detection}, author = {Georgios Bouchouras, Konstantinos Kotis}, url = {https://www.mdpi.com/1999-4893/18/1/34}, doi = {https://doi.org/10.3390/a18010034}, year = {2025}, date = {2025-01-09}, journal = {Algorithms}, volume = {18}, number = {1}, pages = {34}, abstract = {This paper presents a systematic review of the emerging applications of artificial intelligence (AI), Internet of Things (IoT), and sensor-based technologies in the diagnosis of autism spectrum disorder (ASD). The integration of these technologies has led to promising advances in identifying unique behavioral, physiological, and neuroanatomical markers associated with ASD. Through an examination of recent studies, we explore how technologies such as wearable sensors, eye-tracking systems, virtual reality environments, neuroimaging, and microbiome analysis contribute to a holistic approach to ASD diagnostics. The analysis reveals how these technologies facilitate non-invasive, real-time assessments across diverse settings, enhancing both diagnostic accuracy and accessibility. The findings underscore the transformative potential of AI, IoT, and sensor-based driven tools in providing personalized and continuous ASD detection, advocating for data-driven approaches that extend beyond traditional methodologies. Ultimately, this review emphasizes the role of technology in improving ASD diagnostic processes, paving the way for targeted and individualized assessments.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This paper presents a systematic review of the emerging applications of artificial intelligence (AI), Internet of Things (IoT), and sensor-based technologies in the diagnosis of autism spectrum disorder (ASD). The integration of these technologies has led to promising advances in identifying unique behavioral, physiological, and neuroanatomical markers associated with ASD. Through an examination of recent studies, we explore how technologies such as wearable sensors, eye-tracking systems, virtual reality environments, neuroimaging, and microbiome analysis contribute to a holistic approach to ASD diagnostics. The analysis reveals how these technologies facilitate non-invasive, real-time assessments across diverse settings, enhancing both diagnostic accuracy and accessibility. The findings underscore the transformative potential of AI, IoT, and sensor-based driven tools in providing personalized and continuous ASD detection, advocating for data-driven approaches that extend beyond traditional methodologies. Ultimately, this review emphasizes the role of technology in improving ASD diagnostic processes, paving the way for targeted and individualized assessments. |
2024 |
Georgios Bouchouras Georgios Sofianidis, Konstantinos Kotis Temporal Anomaly Detection in Attention-Deficit/Hyperactivity Disorder Using Recurrent Neural Networks Journal Article Cureus, 16 (12), 2024. @article{Bouchouras2024b, title = {Temporal Anomaly Detection in Attention-Deficit/Hyperactivity Disorder Using Recurrent Neural Networks}, author = {Georgios Bouchouras, Georgios Sofianidis, Konstantinos Kotis}, url = {https://assets.cureus.com/uploads/original_article/pdf/324320/20250127-834463-ohpyph.pdf}, doi = {https://doi.org/10.7759/cureus.76496}, year = {2024}, date = {2024-12-27}, journal = {Cureus}, volume = {16}, number = {12}, abstract = {Attention-deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental condition marked by movement hyperactivity, often persisting into adulthood. Understanding the movement patterns associated with ADHD is crucial for improving diagnostic precision and tailoring interventions. This study leverages the HYPERAKTIV dataset, which includes high-resolution temporal data on motor activity from people diagnosed with ADHD. We used the isolation forest algorithm to detect anomalies in activity data, followed by the development of a recurrent neural network (RNN) model to predict these anomalies over time. The RNN model demonstrated high predictive accuracy, with a mean accuracy of 0.953 and a mean loss of 0.124 for participants with ADHD. These findings suggest that machine learning techniques, particularly RNNs, can effectively identify and predict anomalies in temporal motor activity data, offering objective insights into ADHD-related movement behaviors. This approach is promising for informing personalized interventions and improving clinical decision-making in the management of ADHD.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Attention-deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental condition marked by movement hyperactivity, often persisting into adulthood. Understanding the movement patterns associated with ADHD is crucial for improving diagnostic precision and tailoring interventions. This study leverages the HYPERAKTIV dataset, which includes high-resolution temporal data on motor activity from people diagnosed with ADHD. We used the isolation forest algorithm to detect anomalies in activity data, followed by the development of a recurrent neural network (RNN) model to predict these anomalies over time. The RNN model demonstrated high predictive accuracy, with a mean accuracy of 0.953 and a mean loss of 0.124 for participants with ADHD. These findings suggest that machine learning techniques, particularly RNNs, can effectively identify and predict anomalies in temporal motor activity data, offering objective insights into ADHD-related movement behaviors. This approach is promising for informing personalized interventions and improving clinical decision-making in the management of ADHD. |
Davide Ferraris Konstantinos Kotis, Christos Kalloniatis Enhancing TrUStAPIS Methodology in the Web of Things with LLM-generated IoT Trust Semantics Conference The 2024 International Conference on Information and Communications Security (ICICS 2024), 2024. @conference{Ferraris2024, title = {Enhancing TrUStAPIS Methodology in the Web of Things with LLM-generated IoT Trust Semantics}, author = {Davide Ferraris, Konstantinos Kotis, Christos Kalloniatis}, url = {https://link.springer.com/chapter/10.1007/978-981-97-8798-2_7}, doi = {https://doi.org/10.1007/978-981-97-8798-2_7}, year = {2024}, date = {2024-12-25}, booktitle = {The 2024 International Conference on Information and Communications Security (ICICS 2024)}, pages = {125-144}, abstract = {In the Internet of Things (IoT) there are ecosystems where their physical ’smart’ entities virtually interact with each other. Often, this interaction occurs among unknown entities, making trust an essential requirement to overcome uncertainty in several aspects of this interaction. However, trust is a complex concept, and incorporating it in IoT is still a challenging topic. For this reason, it is highly significant to specify and model trust in early stages of the System Development Life Cycle (SDLC) of IoT-integrated systems, thus enhancing the aforementioned task. TrUStAPIS is a requirements engineering methodology recently introduced for incorporating trust requirements during IoT-based system design. The scope of this paper is to provide an extension of TrUStAPIS by introducing IoT trust semantics compatible with the W3C Web of Things (WoT) recommendations generated with the assistance of Large Language Models (LLMs). Taking advantage of LLMs as a tool for integrating and refining existing methodologies, in this paper we present our work towards a revision of the TrUStAPIS methodology. In this work, we contribute a new conceptual model and a refined JSON-LD ontology that takes into account IoT trust semantics, providing eventually a valuable tool for software engineers to design and model IoT-based systems and services.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } In the Internet of Things (IoT) there are ecosystems where their physical ’smart’ entities virtually interact with each other. Often, this interaction occurs among unknown entities, making trust an essential requirement to overcome uncertainty in several aspects of this interaction. However, trust is a complex concept, and incorporating it in IoT is still a challenging topic. For this reason, it is highly significant to specify and model trust in early stages of the System Development Life Cycle (SDLC) of IoT-integrated systems, thus enhancing the aforementioned task. TrUStAPIS is a requirements engineering methodology recently introduced for incorporating trust requirements during IoT-based system design. The scope of this paper is to provide an extension of TrUStAPIS by introducing IoT trust semantics compatible with the W3C Web of Things (WoT) recommendations generated with the assistance of Large Language Models (LLMs). Taking advantage of LLMs as a tool for integrating and refining existing methodologies, in this paper we present our work towards a revision of the TrUStAPIS methodology. In this work, we contribute a new conceptual model and a refined JSON-LD ontology that takes into account IoT trust semantics, providing eventually a valuable tool for software engineers to design and model IoT-based systems and services. |
Andreas Soularidis Konstantinos Kotis, Myriam Lamolle Zakaria Mejdoul Gaëlle Lortal George Vouros LLM-Assisted Generation of SWRL Rules from Natural Language Conference 2024 International Conference on AI x Data and Knowledge Engineering (AIxDKE), 2024, ISBN: 979-8-3315-1704-5. @conference{Soularidis2024b, title = {LLM-Assisted Generation of SWRL Rules from Natural Language}, author = {Andreas Soularidis, Konstantinos Kotis, Myriam Lamolle, Zakaria Mejdoul, Gaëlle Lortal, George Vouros}, doi = {https://doi.org/10.1109/AIxDKE63520.2024.00008}, isbn = {979-8-3315-1704-5}, year = {2024}, date = {2024-12-11}, booktitle = {2024 International Conference on AI x Data and Knowledge Engineering (AIxDKE)}, abstract = {Recently, Large Language Models (LLMs) have attracted great attention due to their remarkable performance in human-like text generation and reasoning skills (although their memory and hallucination problems still remain key issues to tackle more efficiently). LLMs have been applied to various application domains, including Knowledge Graph (KG) generation, question and answering over KGs and text-to-SPARQL translation. In this work, we investigate the capabilities of LLMs in text-to-SWRL translation, i.e., translation of Natural Language (NL) rules into Semantic Web Rule Language (SWRL) rules, put in the context of an industrial Ontology Engineering (OE) environment called GLUON, presenting our first experimental results. The aim of this work is to identify the level of automation that is adequate for the LLM to generate well-formed SWRL rules, towards the development of an LLM-based framework, as a plugin to the GLUON OE environment. In this direction we leverage and combine the reasoning capabilities of GPT-4o model, the Retrieval-Augmented Generation (RAG) technology, and prompt engineering. We employ quantitative and qualitative metrics to evaluate the generated SWRL rules, focusing on the correct syntax and the level of human intervention.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } Recently, Large Language Models (LLMs) have attracted great attention due to their remarkable performance in human-like text generation and reasoning skills (although their memory and hallucination problems still remain key issues to tackle more efficiently). LLMs have been applied to various application domains, including Knowledge Graph (KG) generation, question and answering over KGs and text-to-SPARQL translation. In this work, we investigate the capabilities of LLMs in text-to-SWRL translation, i.e., translation of Natural Language (NL) rules into Semantic Web Rule Language (SWRL) rules, put in the context of an industrial Ontology Engineering (OE) environment called GLUON, presenting our first experimental results. The aim of this work is to identify the level of automation that is adequate for the LLM to generate well-formed SWRL rules, towards the development of an LLM-based framework, as a plugin to the GLUON OE environment. In this direction we leverage and combine the reasoning capabilities of GPT-4o model, the Retrieval-Augmented Generation (RAG) technology, and prompt engineering. We employ quantitative and qualitative metrics to evaluate the generated SWRL rules, focusing on the correct syntax and the level of human intervention. |
Francesk Mulita Chrysa Apostoloumi, Admir Mulita Georgios-Ioannis Verras Michail Pitiakoudis Konstantinos Kotis Christos-Nikolaos Anagnostopoulos The use of artificial intelligence in surgical oncology simulation Journal Article European Journal of Surgical Oncology, 50 , 2024. @article{Mulita2024, title = {The use of artificial intelligence in surgical oncology simulation}, author = {Francesk Mulita, Chrysa Apostoloumi, Admir Mulita, Georgios-Ioannis Verras, Michail Pitiakoudis, Konstantinos Kotis, Christos-Nikolaos Anagnostopoulos}, url = {https://www.sciencedirect.com/science/article/abs/pii/S0748798324015063}, doi = {https://doi.org/10.1016/j.ejso.2024.109438}, year = {2024}, date = {2024-12-01}, journal = {European Journal of Surgical Oncology}, volume = {50}, abstract = {We systematically searched PubMed from inception to 10 July 2024. Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines were followed for this systematic review. Keywords used for search were (“Artificial intelligence”) AND (“Surgical education” OR “Surgical training”). The search was conducted on 5 August 2024.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We systematically searched PubMed from inception to 10 July 2024. Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines were followed for this systematic review. Keywords used for search were (“Artificial intelligence”) AND (“Surgical education” OR “Surgical training”). The search was conducted on 5 August 2024. |
Admir Mulita Francesk Mulita, Georgios-Ioannis Verras Konstantinos Kotis Christos-Nikolaos Anagnostopoulos Using Deep Learning to Predict Head-Neck and Lung Cancer Treatment Response from Serial Medical Imaging Journal Article European Journal of Surgical Oncology, 50 , pp. 109437, 2024, ISSN: 0748-7983. @article{Mulita2024b, title = {Using Deep Learning to Predict Head-Neck and Lung Cancer Treatment Response from Serial Medical Imaging}, author = {Admir Mulita, Francesk Mulita, Georgios-Ioannis Verras, Konstantinos Kotis, Christos-Nikolaos Anagnostopoulos}, doi = {https://doi.org/10.1016/j.ejso.2024.109437}, issn = {0748-7983}, year = {2024}, date = {2024-12-01}, journal = {European Journal of Surgical Oncology}, volume = {50}, pages = {109437}, abstract = {Tumors are continuously evolving biological systems, and medical imaging is uniquely positioned to monitor changes throughout treatment. Although qualitatively tracking lesions over space and time may be trivial, the development of clinically relevant, automated radiomics methods that incorporate serial imaging data is far more challenging. This review evaluates deep learning networks for predicting clinical outcomes through analyzing time series Positron Emission tomography – PET and CT images of patients with head -neck and lung cancer.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Tumors are continuously evolving biological systems, and medical imaging is uniquely positioned to monitor changes throughout treatment. Although qualitatively tracking lesions over space and time may be trivial, the development of clinically relevant, automated radiomics methods that incorporate serial imaging data is far more challenging. This review evaluates deep learning networks for predicting clinical outcomes through analyzing time series Positron Emission tomography – PET and CT images of patients with head -neck and lung cancer. |
Andreas Sideras Konstantinos Bougiatiotis, Elias Zavitsanos Georgios Paliouras George Vouros Bankruptcy Prediction: Data Augmentation, LLMs and the Need for Auditor’s Opinion Conference ICAIF ’24: Proceedings of the 5th ACM International Conference on AI in Finance, 2024. @conference{Sideras2024, title = {Bankruptcy Prediction: Data Augmentation, LLMs and the Need for Auditor’s Opinion}, author = {Andreas Sideras, Konstantinos Bougiatiotis, Elias Zavitsanos, Georgios Paliouras, George Vouros}, url = {https://dl.acm.org/doi/epdf/10.1145/3677052.3698627}, doi = {https://doi.org/10.1145/3677052.3698627}, year = {2024}, date = {2024-11-14}, booktitle = {ICAIF ’24: Proceedings of the 5th ACM International Conference on AI in Finance}, abstract = {Predicting bankruptcy is crucial for managing financial risk in corporations. This study emphasizes incorporating the auditor’s opinion text into prediction models to improve their ability to assess financial health. These opinions provide essential insights as they offer an independent assessment, complementing other predictive inputs like the management’s discussion and analysis. However, the rarity of bankruptcy cases in the data introduces a challenging issue due to severe class imbalance. To address this, we propose a method to generate synthetic positive samples using a variational autoencoder and integrate the multi-source input in a late fusion setting. We showcase that both data augmentation and using multiple textual sources improve the performance of existing models on a related benchmark dataset. Additionally, we evaluate LLMs when used for data augmentation in the proposed method and in a zero-shot prediction setting, discussing important aspects to consider when incorporating them in a predictive pipeline.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } Predicting bankruptcy is crucial for managing financial risk in corporations. This study emphasizes incorporating the auditor’s opinion text into prediction models to improve their ability to assess financial health. These opinions provide essential insights as they offer an independent assessment, complementing other predictive inputs like the management’s discussion and analysis. However, the rarity of bankruptcy cases in the data introduces a challenging issue due to severe class imbalance. To address this, we propose a method to generate synthetic positive samples using a variational autoencoder and integrate the multi-source input in a late fusion setting. We showcase that both data augmentation and using multiple textual sources improve the performance of existing models on a related benchmark dataset. Additionally, we evaluate LLMs when used for data augmentation in the proposed method and in a zero-shot prediction setting, discussing important aspects to consider when incorporating them in a predictive pipeline. |
George Vouros Ioannis Ioannidis, Georgios Santipantakis Theodore Tranos Konstantinos Blekas Marc Melgosa Xavier Prats Machine-learning methods estimating flights’ hidden parameters for the prediction of KPIs Journal Article Aerospace, 11 (11), pp. 937, 2024. @article{Vouros2024, title = {Machine-learning methods estimating flights’ hidden parameters for the prediction of KPIs}, author = {George Vouros, Ioannis Ioannidis, Georgios Santipantakis, Theodore Tranos, Konstantinos Blekas, Marc Melgosa, Xavier Prats}, url = {https://www.mdpi.com/2226-4310/11/11/937}, doi = {https://doi.org/10.3390/aerospace11110937}, year = {2024}, date = {2024-11-12}, journal = {Aerospace}, volume = {11}, number = {11}, pages = {937}, abstract = {Complex microscopic simulation models of strategic Air Traffic Management (ATM) performance assessment and decision-making are hindered by several factors. One of the most important is the existence of hidden parameters—such as aircraft take-off weight (TOW) and the selected cost index (CI)—which, if known, would allow for more effective performance modeling methodologies for assessing Key Performance Indicators (KPIs) at various levels of abstraction/detail, e.g., system-wide, or at the level of individual flights. This research proposes a data-driven methodology for the estimation of flights’ hidden parameters combining mechanistic and advanced Artificial Intelligence/Machine Learning (AI/ML) models. Aiming at microsimulation models, our goal is to study the effect of these estimations on the prediction of flights’ KPIs. In so doing, we propose a novel methodology according to which data-driven methods are trained given optimal trajectories (produced by mechanistic models) corresponding to known hidden parameter values, with the aim of predicting hidden parameters’ values of unseen trajectories. The results show that estimations of hidden parameters support the accurate prediction of KPIs regarding the efficiency of flights: fuel consumption, gate-to-gate time and distance flown.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Complex microscopic simulation models of strategic Air Traffic Management (ATM) performance assessment and decision-making are hindered by several factors. One of the most important is the existence of hidden parameters—such as aircraft take-off weight (TOW) and the selected cost index (CI)—which, if known, would allow for more effective performance modeling methodologies for assessing Key Performance Indicators (KPIs) at various levels of abstraction/detail, e.g., system-wide, or at the level of individual flights. This research proposes a data-driven methodology for the estimation of flights’ hidden parameters combining mechanistic and advanced Artificial Intelligence/Machine Learning (AI/ML) models. Aiming at microsimulation models, our goal is to study the effect of these estimations on the prediction of flights’ KPIs. In so doing, we propose a novel methodology according to which data-driven methods are trained given optimal trajectories (produced by mechanistic models) corresponding to known hidden parameter values, with the aim of predicting hidden parameters’ values of unseen trajectories. The results show that estimations of hidden parameters support the accurate prediction of KPIs regarding the efficiency of flights: fuel consumption, gate-to-gate time and distance flown. |
Fotis Assimakopoulos Costas Vassilakis, Dionisis Margaris Konstantinos Kotis Dimitris Spiliotopoulos Artificial intelligence tools for the agriculture value chain: Status and prospects Journal Article Electronics, 13 (22), pp. 4362, 2024. @article{Assimakopoulos2024b, title = {Artificial intelligence tools for the agriculture value chain: Status and prospects}, author = {Fotis Assimakopoulos, Costas Vassilakis, Dionisis Margaris, Konstantinos Kotis, Dimitris Spiliotopoulos}, url = {https://www.mdpi.com/2079-9292/13/22/4362}, doi = {https://doi.org/10.3390/electronics13224362}, year = {2024}, date = {2024-11-07}, journal = {Electronics}, volume = {13}, number = {22}, pages = {4362}, abstract = {This article explores the transformative potential of artificial intelligence (AI) tools across the agricultural value chain, highlighting their applications, benefits, challenges, and future prospects. With global food demand projected to increase by 70% by 2050, AI technologies—including machine learning, big data analytics, and the Internet of things (IoT)—offer critical solutions for enhancing agricultural productivity, sustainability, and resource efficiency. The study provides a comprehensive review of AI applications at multiple stages of the agricultural value chain, including land use planning, crop selection, resource management, disease detection, yield prediction, and market integration. It also discusses the significant challenges to AI adoption, such as data accessibility, technological infrastructure, and the need for specialized skills. By examining case studies and empirical evidence, the article demonstrates how AI-driven solutions can optimize decision-making and operational efficiency in agriculture. The findings underscore AI’s pivotal role in addressing global agricultural challenges, with implications for farmers, agribusinesses, policymakers, and researchers. This article aims to advance the evolving research and discussions on sustainable agriculture, contributing insights that promote the adoption of AI technologies and influence the future of farming.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This article explores the transformative potential of artificial intelligence (AI) tools across the agricultural value chain, highlighting their applications, benefits, challenges, and future prospects. With global food demand projected to increase by 70% by 2050, AI technologies—including machine learning, big data analytics, and the Internet of things (IoT)—offer critical solutions for enhancing agricultural productivity, sustainability, and resource efficiency. The study provides a comprehensive review of AI applications at multiple stages of the agricultural value chain, including land use planning, crop selection, resource management, disease detection, yield prediction, and market integration. It also discusses the significant challenges to AI adoption, such as data accessibility, technological infrastructure, and the need for specialized skills. By examining case studies and empirical evidence, the article demonstrates how AI-driven solutions can optimize decision-making and operational efficiency in agriculture. The findings underscore AI’s pivotal role in addressing global agricultural challenges, with implications for farmers, agribusinesses, policymakers, and researchers. This article aims to advance the evolving research and discussions on sustainable agriculture, contributing insights that promote the adoption of AI technologies and influence the future of farming. |
Alexandros Troupiotis-Kapeliaris Nikolaos Sapountzis, Giannis Spiliopoulos Thomas Kogias Piyabhum Chaysri Bernardo AntÓnio Correia Gabriel Miguel Filipe Santos Silva Elias Xidias Konstantinos Blekas Dimitris Zissis Collection: The Aegean Ro-Boat Race 2023 Journal Article IEEE Data Descriptions, 1 , pp. 87-94, 2024, ISSN: 2995-4274. @article{Troupiotis-Kapeliaris2024, title = {Collection: The Aegean Ro-Boat Race 2023}, author = {Alexandros Troupiotis-Kapeliaris, Nikolaos Sapountzis, Giannis Spiliopoulos, Thomas Kogias, Piyabhum Chaysri, Bernardo AntÓnio Correia Gabriel, Miguel Filipe Santos Silva, Elias Xidias, Konstantinos Blekas, Dimitris Zissis}, url = {https://ieeexplore.ieee.org/abstract/document/10707278}, doi = {https://doi.org/10.1109/IEEEDATA.2024.3475332}, issn = {2995-4274}, year = {2024}, date = {2024-10-08}, journal = {IEEE Data Descriptions}, volume = {1}, pages = {87-94}, abstract = {In this article, we introduce a publicly available real-world dataset collected during the Aegean Ro-Boat Race 2023, which took place at the University of the Aegean in Syros, Greece. The Aegean Ro-Boat Race represents an international competition at the university level, challenging teams to innovate and develop autonomous marine robotic systems capable of performing in unknown dynamic maritime environments under real-world conditions. The 2023 competition featured three primary mission tasks, each designed to test different aspects of the robotic systems: 1) high-speed performance for evaluating the speed and agility of the autonomous vessels; 2) collision avoidance for assessing the systems’ ability to detect and avoid obstacles in real-time; and 3) endurance for testing the operational longevity and efficiency of the robotic systems over extended periods. In total, seven teams registered for the competition, with five of them being from Greece and two from the countries of Portugal and Latvia. Due to several technical difficulties, three vessels were able to complete all races, and data were recorded during their entire participation. The spatiotemporal data for the “Aegean Ro-Boat Race” was gathered through an onboard data logging system that continuously monitored various sensors, including global positioning system (GPS), for all vessels during the entire competition. The dataset includes positional reports from the vessels during all three races (totaling over 6500 records), the positions of the external track and obstacle buoys, together with a file regarding the weather conditions during the race day.}, keywords = {}, pubstate = {published}, tppubtype = {article} } In this article, we introduce a publicly available real-world dataset collected during the Aegean Ro-Boat Race 2023, which took place at the University of the Aegean in Syros, Greece. The Aegean Ro-Boat Race represents an international competition at the university level, challenging teams to innovate and develop autonomous marine robotic systems capable of performing in unknown dynamic maritime environments under real-world conditions. The 2023 competition featured three primary mission tasks, each designed to test different aspects of the robotic systems: 1) high-speed performance for evaluating the speed and agility of the autonomous vessels; 2) collision avoidance for assessing the systems’ ability to detect and avoid obstacles in real-time; and 3) endurance for testing the operational longevity and efficiency of the robotic systems over extended periods. In total, seven teams registered for the competition, with five of them being from Greece and two from the countries of Portugal and Latvia. Due to several technical difficulties, three vessels were able to complete all races, and data were recorded during their entire participation. The spatiotemporal data for the “Aegean Ro-Boat Race” was gathered through an onboard data logging system that continuously monitored various sensors, including global positioning system (GPS), for all vessels during the entire competition. The dataset includes positional reports from the vessels during all three races (totaling over 6500 records), the positions of the external track and obstacle buoys, together with a file regarding the weather conditions during the race day. |
Theodore Tranos Piyabhum Chaysri, Christos Spatharis Konstantinos Blekas SETN ’24: Proceedings of the 13th Hellenic Conference on Artificial Intelligence , 2024. @conference{Tranos2024b, title = {An Advanced Deep Reinforcement Learning Framework for Docking Unmanned Surface Vessels in Variable Environmental Conditions and Amid Moving Ships}, author = {Theodore Tranos, Piyabhum Chaysri, Christos Spatharis, Konstantinos Blekas}, url = {https://dl.acm.org/doi/full/10.1145/3688671.3688779}, doi = {https://doi.org/10.1145/3688671.3688779}, year = {2024}, date = {2024-09-11}, booktitle = {SETN ’24: Proceedings of the 13th Hellenic Conference on Artificial Intelligence }, pages = {1-10}, abstract = {This work utilizes advanced reinforcement learning techniques to optimize the docking process of unmanned surface vessels (USVs) in challenging maritime environments. The proposed methodology accounts for fluctuating environmental conditions, such as wind, as well as the dynamic presence of moving ships as objects and varying traffic densities. It integrates learning models and rich state spaces with predictive information to enhance decision-making capabilities that can anticipate and react to environmental changes and vessel movements, ensuring safe and efficient docking procedures. Experiments were conducted in a simulated environment of the port of Piraeus using real data on environmental conditions and ship movements. In addition, we tried to evaluate the proposed method in conditions different from those it was trained, in order to measure the generalization ability of the USV agent’s policy and also to achieve more plausibility in the results.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } This work utilizes advanced reinforcement learning techniques to optimize the docking process of unmanned surface vessels (USVs) in challenging maritime environments. The proposed methodology accounts for fluctuating environmental conditions, such as wind, as well as the dynamic presence of moving ships as objects and varying traffic densities. It integrates learning models and rich state spaces with predictive information to enhance decision-making capabilities that can anticipate and react to environmental changes and vessel movements, ensuring safe and efficient docking procedures. Experiments were conducted in a simulated environment of the port of Piraeus using real data on environmental conditions and ship movements. In addition, we tried to evaluate the proposed method in conditions different from those it was trained, in order to measure the generalization ability of the USV agent’s policy and also to achieve more plausibility in the results. |
Fotis Assimakopoulos Costas Vassilakis, Dionisis Margaris Konstantinos Kotis Dimitris Spiliotopoulos The implementation of “smart” technologies in the agricultural sector: a review Journal Article Information, 15 (8), pp. 466, 2024. @article{Assimakopoulos2024, title = {The implementation of “smart” technologies in the agricultural sector: a review}, author = {Fotis Assimakopoulos, Costas Vassilakis, Dionisis Margaris, Konstantinos Kotis, Dimitris Spiliotopoulos}, url = {https://www.mdpi.com/2078-2489/15/8/466}, doi = {https://doi.org/10.3390/info15080466}, year = {2024}, date = {2024-08-06}, journal = {Information}, volume = {15}, number = {8}, pages = {466}, abstract = {The growing global population demands an increase in agricultural production and the promotion of sustainable practices. Smart agriculture, driven by advanced technologies, is crucial to achieving these goals. These technologies provide real-time information for crop monitoring, yield prediction, and essential farming functions. However, adopting intelligent farming systems poses challenges, including learning new systems and dealing with installation costs. Robust support is crucial for integrating smart farming into practices. Understanding the current state of agriculture, technology trends, and the challenges in technology acceptance is essential for a smooth transition to Agriculture 4.0. This work reports on the pivotal synergy of IoT technology with other research trends, such as weather forecasting and robotics. It also presents the applications of smart agriculture worldwide, with an emphasis on government initiatives to support farmers and promote global adoption. The aim of this work is to provide a comprehensive review of smart technologies for precision agriculture and especially of their adoption level and results on the global scale; to this end, this review examines three important areas of smart agriculture, namely field, greenhouse, and livestock monitoring.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The growing global population demands an increase in agricultural production and the promotion of sustainable practices. Smart agriculture, driven by advanced technologies, is crucial to achieving these goals. These technologies provide real-time information for crop monitoring, yield prediction, and essential farming functions. However, adopting intelligent farming systems poses challenges, including learning new systems and dealing with installation costs. Robust support is crucial for integrating smart farming into practices. Understanding the current state of agriculture, technology trends, and the challenges in technology acceptance is essential for a smooth transition to Agriculture 4.0. This work reports on the pivotal synergy of IoT technology with other research trends, such as weather forecasting and robotics. It also presents the applications of smart agriculture worldwide, with an emphasis on government initiatives to support farmers and promote global adoption. The aim of this work is to provide a comprehensive review of smart technologies for precision agriculture and especially of their adoption level and results on the global scale; to this end, this review examines three important areas of smart agriculture, namely field, greenhouse, and livestock monitoring. |
Theodore Tranos Christos Spatharis, Konstantinos Blekas Andreas-Giorgios Stafylopatis Large-Scale Urban Traffic Management Using Zero-Shot Knowledge Transfer in Multi-Agent Reinforcement Learning for Intersection Patterns Journal Article Robotics, 13 (7), pp. 109, 2024. @article{Tranos2024, title = {Large-Scale Urban Traffic Management Using Zero-Shot Knowledge Transfer in Multi-Agent Reinforcement Learning for Intersection Patterns}, author = {Theodore Tranos, Christos Spatharis, Konstantinos Blekas, Andreas-Giorgios Stafylopatis}, url = {https://www.mdpi.com/2218-6581/13/7/109}, doi = {https://doi.org/10.3390/robotics13070109}, year = {2024}, date = {2024-07-19}, journal = {Robotics}, volume = {13}, number = {7}, pages = {109}, abstract = {The automatic control of vehicle traffic in large urban networks constitutes one of the most serious challenges to modern societies, with an impact on improving the quality of human life and saving energy and time. Intersections are a special traffic structure of pivotal importance as they accumulate a large number of vehicles that should be served in an optimal manner. Constructing intelligent models that manage to automatically coordinate and steer vehicles through intersections is a key point in the fragmentation of traffic control, offering active solutions through the flexibility of automatically adapting to a variety of traffic conditions. Responding to this call, this work aims to propose an integrated active solution of automatic traffic management. We introduce a multi-agent reinforcement learning framework that effectively models traffic flow at individual unsignalized intersections. It relies on a compact agent definition, a rich information state space, and a learning process characterized not only by depth and quality, but also by substantial degrees of freedom and variability. The resulting driving profiles are further transferred to larger road networks to integrate their individual elements and compose an effective automatic traffic control platform. Experiments are conducted on simulated road networks of variable complexity, demonstrating the potential of the proposed method.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The automatic control of vehicle traffic in large urban networks constitutes one of the most serious challenges to modern societies, with an impact on improving the quality of human life and saving energy and time. Intersections are a special traffic structure of pivotal importance as they accumulate a large number of vehicles that should be served in an optimal manner. Constructing intelligent models that manage to automatically coordinate and steer vehicles through intersections is a key point in the fragmentation of traffic control, offering active solutions through the flexibility of automatically adapting to a variety of traffic conditions. Responding to this call, this work aims to propose an integrated active solution of automatic traffic management. We introduce a multi-agent reinforcement learning framework that effectively models traffic flow at individual unsignalized intersections. It relies on a compact agent definition, a rich information state space, and a learning process characterized not only by depth and quality, but also by substantial degrees of freedom and variability. The resulting driving profiles are further transferred to larger road networks to integrate their individual elements and compose an effective automatic traffic control platform. Experiments are conducted on simulated road networks of variable complexity, demonstrating the potential of the proposed method. |
Sideras, Andreas Multimodal pretraining for music audio Masters Thesis Department of Digital Systems, School of Information Technologies and Communication, University of Piraeus, 2024. @mastersthesis{Sideras2024b, title = {Multimodal pretraining for music audio}, author = {Andreas Sideras}, url = {https://dione.lib.unipi.gr/xmlui/handle/unipi/16697 https://dione.lib.unipi.gr/xmlui/bitstream/handle/unipi/16697/Andreas-Sideras-MSc-Thesis.pdf?sequence=1&isAllowed=y}, doi = {http://dx.doi.org/10.26267/unipi_dione/4119}, year = {2024}, date = {2024-07-01}, school = {Department of Digital Systems, School of Information Technologies and Communication, University of Piraeus}, abstract = {Data can be expressed in various forms, each potentially encoded through diverse means. For instance, we might encounter audio data paired with descriptive texts about their lyrics. Modern systems leverage, if available, the different sources of information and outperform, under certain conditions, their single-modal counterparts. In such multimodal settings, each modality encapsulates a distinct aspect of the underlying semantics of the data and has a supplementary role. Data can also be limited and without annotations related to the task at hand. In such cases, transfer learning and pretraining could be two techniques that enhance the performance of the models. In this thesis, we explore various unsupervised pretraining techniques while evaluating them on a supervised downstream task. Our goal is to train a model that can extract meaningful features and be further finetuned to any new task. We use LLMs to create pseudo-captions that describe the sentiment and the theme of the lyrics, from a large pool of non-annotated audio. We then perform a pretraining step, where we learn a multimodal coordinated space between the audio signals and these pseudo-captions. Then, we finetune our model on an annotated dataset, where only the audio modality is available. We highlight the ability of such models to deliver adequate performance in few-shot learning settings, the incorporation of LLMs into the pretraining step, and the importance of learning a shared semantic space for information originating from different modalities.}, keywords = {}, pubstate = {published}, tppubtype = {mastersthesis} } Data can be expressed in various forms, each potentially encoded through diverse means. For instance, we might encounter audio data paired with descriptive texts about their lyrics. Modern systems leverage, if available, the different sources of information and outperform, under certain conditions, their single-modal counterparts. In such multimodal settings, each modality encapsulates a distinct aspect of the underlying semantics of the data and has a supplementary role. Data can also be limited and without annotations related to the task at hand. In such cases, transfer learning and pretraining could be two techniques that enhance the performance of the models. In this thesis, we explore various unsupervised pretraining techniques while evaluating them on a supervised downstream task. Our goal is to train a model that can extract meaningful features and be further finetuned to any new task. We use LLMs to create pseudo-captions that describe the sentiment and the theme of the lyrics, from a large pool of non-annotated audio. We then perform a pretraining step, where we learn a multimodal coordinated space between the audio signals and these pseudo-captions. Then, we finetune our model on an annotated dataset, where only the audio modality is available. We highlight the ability of such models to deliver adequate performance in few-shot learning settings, the incorporation of LLMs into the pretraining step, and the importance of learning a shared semantic space for information originating from different modalities. |
Mariana Ziku Konstantinos Kotis, Gerasimos Pavlogeorgatos Evangelia Kavakli Chara Zeeri George Caridakis Evaluating crowdsourcing applications with map-based storytelling capabilities in cultural heritage Journal Article Heritage, 7 (7), pp. 3429-3454, 2024. @article{Ziku2024, title = {Evaluating crowdsourcing applications with map-based storytelling capabilities in cultural heritage}, author = {Mariana Ziku, Konstantinos Kotis, Gerasimos Pavlogeorgatos, Evangelia Kavakli, Chara Zeeri, George Caridakis}, url = {https://www.mdpi.com/2571-9408/7/7/162}, doi = {https://doi.org/10.3390/heritage7070162}, year = {2024}, date = {2024-06-28}, journal = {Heritage}, volume = {7}, number = {7}, pages = {3429-3454}, abstract = {Crowdsourcing applications that integrate storytelling and geotagging capabilities offer new avenues for engaging the public in cultural heritage. However, standardised evaluation frameworks are lacking. This paper presents an applied evaluation methodology involving the analysis of relevant web-based tools. Towards this goal, this paper presents the development of crowdsourcing applications using, as a case study, the collection of myths and legends associated with the monumental heritage site of the Palace of the Grand Master of the Knights of Rhodes in Greece. Additionally, the paper presents an evaluation conducted through a criteria-based approach and user-based survey. The study reviews the concepts of crowdsourcing and digital storytelling within digital heritage, along with current concepts of living heritage and folklore, and examines relevant initiatives. The evaluation follows a four-stage methodology: (i) initial web-based tool selection based on the minimum requirements of web compatibility, crowdsourced data display, and map-based storytelling capability; (ii) application development using the selected web-based tools; (iii) a five-criteria assessment, based on the factors of open access, usability/tool support, participatory content/story creation, metrics provision and metadata model usage; and (iv) a crowd-based survey, indicating the most effective option. Findings from 100 respondents reveal limited exposure to participatory storytelling applications but interest in contributing content. Social media and influential figures serve as key channels for promoting crowdsourcing open calls. The results highlight gaps in understanding user expectations and perceptions, suggesting future research for gaining insights into engagement rates.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Crowdsourcing applications that integrate storytelling and geotagging capabilities offer new avenues for engaging the public in cultural heritage. However, standardised evaluation frameworks are lacking. This paper presents an applied evaluation methodology involving the analysis of relevant web-based tools. Towards this goal, this paper presents the development of crowdsourcing applications using, as a case study, the collection of myths and legends associated with the monumental heritage site of the Palace of the Grand Master of the Knights of Rhodes in Greece. Additionally, the paper presents an evaluation conducted through a criteria-based approach and user-based survey. The study reviews the concepts of crowdsourcing and digital storytelling within digital heritage, along with current concepts of living heritage and folklore, and examines relevant initiatives. The evaluation follows a four-stage methodology: (i) initial web-based tool selection based on the minimum requirements of web compatibility, crowdsourced data display, and map-based storytelling capability; (ii) application development using the selected web-based tools; (iii) a five-criteria assessment, based on the factors of open access, usability/tool support, participatory content/story creation, metrics provision and metadata model usage; and (iv) a crowd-based survey, indicating the most effective option. Findings from 100 respondents reveal limited exposure to participatory storytelling applications but interest in contributing content. Social media and influential figures serve as key channels for promoting crowdsourcing open calls. The results highlight gaps in understanding user expectations and perceptions, suggesting future research for gaining insights into engagement rates. |
Dimitrios Doumanas Andreas Soularidis, Konstantinos Kotis George Vouros Integrating LLMs in the Engineering of a SAR Ontology Conference Artificial Intelligence Applications and Innovations, Springer, Cham, 2024, ISBN: 978-3-031-63223-5. @conference{Doumanas2024, title = {Integrating LLMs in the Engineering of a SAR Ontology}, author = {Dimitrios Doumanas, Andreas Soularidis, Konstantinos Kotis, George Vouros}, doi = {https://doi.org/10.1007/978-3-031-63223-5_27}, isbn = {978-3-031-63223-5}, year = {2024}, date = {2024-06-21}, booktitle = { Artificial Intelligence Applications and Innovations}, pages = {360-374}, publisher = {Springer, Cham}, abstract = {In Search and Rescue (SAR) missions, the integration of multiple sources of information may enhance operational efficiency and increase responsiveness significantly, improving situation awareness and aiding decision-making to save lives and mitigate incident impact. Ontologies are crucial for integrating and reasoning with data from diverse sources. Engineering a domain ontology for SAR can be better supported from an agile, collaborative, and iterative ontology engineering methodology (OEM), incorporating the interests of several stakeholders. Large Language Models (LLMs) can play a significant role in completing OEM processes. The goal of this work is to identify how ontology engineering (OE) tasks can be completed with the collaboration of LLMs and humans. The objectives of this paper are, a) to present preliminary exploration of LLMs to generate domain ontologies for the modeling of SAR missions in wildfire incidents b) to propose and evaluate an LLM-enhanced OE approach. In overall, the main contribution of the work presented in this paper is the analysis of LLMs capabilities to ontology engineering, and the evaluation of the synergy between humans and machines to efficiently represent knowledge, with specific focus in the SAR domain.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } In Search and Rescue (SAR) missions, the integration of multiple sources of information may enhance operational efficiency and increase responsiveness significantly, improving situation awareness and aiding decision-making to save lives and mitigate incident impact. Ontologies are crucial for integrating and reasoning with data from diverse sources. Engineering a domain ontology for SAR can be better supported from an agile, collaborative, and iterative ontology engineering methodology (OEM), incorporating the interests of several stakeholders. Large Language Models (LLMs) can play a significant role in completing OEM processes. The goal of this work is to identify how ontology engineering (OE) tasks can be completed with the collaboration of LLMs and humans. The objectives of this paper are, a) to present preliminary exploration of LLMs to generate domain ontologies for the modeling of SAR missions in wildfire incidents b) to propose and evaluate an LLM-enhanced OE approach. In overall, the main contribution of the work presented in this paper is the analysis of LLMs capabilities to ontology engineering, and the evaluation of the synergy between humans and machines to efficiently represent knowledge, with specific focus in the SAR domain. |
Christos Spatharis Konstantinos Blekas, George Vouros A Modelling flight trajectories with multi-modal generative adversarial imitation learning Journal Article Applied Intelligence, 54 (11), pp. 7118-7134, 2024. @article{Spatharis2024c, title = {Modelling flight trajectories with multi-modal generative adversarial imitation learning}, author = {Christos Spatharis, Konstantinos Blekas, George A Vouros}, doi = {https://doi.org/10.1007/s10489-024-05519-6}, year = {2024}, date = {2024-06-03}, journal = {Applied Intelligence}, volume = {54}, number = {11}, pages = {7118-7134}, abstract = {Models of aircraft trajectories become important components of systems supporting the trajectory based operations paradigm: trajectory predictability is considered to be the main driver to enhance operational key performance areas, such as capacity of the airspace, effectiveness regarding all stakeholders’ objectives, and, of course, safety. This article formulates the trajectory modelling problem as a data-driven imitation learning problem addressing multi-modality. To solve this problem we study the use of state-of-the-art multi-modal imitation learning methods Info-GAIL and Triple-GAIL operating in a supervised way, with the aim of (a) disentangling modalities representing patterns of trajectory evolution, and (b) predicting trajectories. Experiments are performed using a real-world dataset of long flights with origin Paris and destination Istanbul. Results show the potential of imitation learning methods to disentangle multi-modal trajectories in real-world settings and predict trajectories with high accuracy.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Models of aircraft trajectories become important components of systems supporting the trajectory based operations paradigm: trajectory predictability is considered to be the main driver to enhance operational key performance areas, such as capacity of the airspace, effectiveness regarding all stakeholders’ objectives, and, of course, safety. This article formulates the trajectory modelling problem as a data-driven imitation learning problem addressing multi-modality. To solve this problem we study the use of state-of-the-art multi-modal imitation learning methods Info-GAIL and Triple-GAIL operating in a supervised way, with the aim of (a) disentangling modalities representing patterns of trajectory evolution, and (b) predicting trajectories. Experiments are performed using a real-world dataset of long flights with origin Paris and destination Istanbul. Results show the potential of imitation learning methods to disentangle multi-modal trajectories in real-world settings and predict trajectories with high accuracy. |
Christos Spatharis Konstantinos Blekas, George Vouros A Modelling flight trajectories with multi-modal generative adversarial imitation learning Journal Article Applied Intelligence, 54 , pp. 7118-7134, 2024. @article{Spatharis2024, title = {Modelling flight trajectories with multi-modal generative adversarial imitation learning}, author = {Christos Spatharis, Konstantinos Blekas, George A. Vouros }, url = {https://link.springer.com/article/10.1007/s10489-024-05519-6}, doi = {https://doi.org/10.1007/s10489-024-05519-6}, year = {2024}, date = {2024-06-03}, journal = {Applied Intelligence}, volume = {54}, pages = {7118-7134}, abstract = {Models of aircraft trajectories become important components of systems supporting the trajectory based operations paradigm: trajectory predictability is considered to be the main driver to enhance operational key performance areas, such as capacity of the airspace, effectiveness regarding all stakeholders’ objectives, and, of course, safety. This article formulates the trajectory modelling problem as a data-driven imitation learning problem addressing multi-modality. To solve this problem we study the use of state-of-the-art multi-modal imitation learning methods Info-GAIL and Triple-GAIL operating in a supervised way, with the aim of (a) disentangling modalities representing patterns of trajectory evolution, and (b) predicting trajectories. Experiments are performed using a real-world dataset of long flights with origin Paris and destination Istanbul. Results show the potential of imitation learning methods to disentangle multi-modal trajectories in real-world settings and predict trajectories with high accuracy.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Models of aircraft trajectories become important components of systems supporting the trajectory based operations paradigm: trajectory predictability is considered to be the main driver to enhance operational key performance areas, such as capacity of the airspace, effectiveness regarding all stakeholders’ objectives, and, of course, safety. This article formulates the trajectory modelling problem as a data-driven imitation learning problem addressing multi-modality. To solve this problem we study the use of state-of-the-art multi-modal imitation learning methods Info-GAIL and Triple-GAIL operating in a supervised way, with the aim of (a) disentangling modalities representing patterns of trajectory evolution, and (b) predicting trajectories. Experiments are performed using a real-world dataset of long flights with origin Paris and destination Istanbul. Results show the potential of imitation learning methods to disentangle multi-modal trajectories in real-world settings and predict trajectories with high accuracy. |
Georgios Bouchouras Pavlos Bitilis, Konstantinos Kotis George Vouros A LLMs for the Engineering of a Parkinson Disease Monitoring and Alerting Ontology. Conference GeNeSy’24: First International Workshop on Generative Neuro-Symbolic Artificial Intelligence, co-located with ESWC 2024, 2024. @conference{Bouchouras2024, title = {LLMs for the Engineering of a Parkinson Disease Monitoring and Alerting Ontology.}, author = {Georgios Bouchouras, Pavlos Bitilis, Konstantinos Kotis, George A Vouros}, url = {https://www.researchgate.net/profile/Giorgos-Bouchouras/publication/383431620_LLMs_for_the_Engineering_of_a_Parkinson_Disease_Monitoring_and_Alerting_Ontology/links/67a859dd4c479b26c9dac1f5/LLMs-for-the-Engineering-of-a-Parkinson-Disease-Monitoring-and-Alerting-Ontology.pdf}, year = {2024}, date = {2024-05-26}, booktitle = {GeNeSy’24: First International Workshop on Generative Neuro-Symbolic Artificial Intelligence, co-located with ESWC 2024}, abstract = {This paper investigates the integration of Large Language Models (LLMs) in the engineering of a Parkinson’s Disease (PD) monitoring and alerting ontology. The focus is on the ontology engineering methodology which combines the capabilities of LLMs and human expertise to develop more robust and comprehensive domain ontologies, faster than humans do alone. Evaluating models like ChatGPT-3.5, ChatGPT4, Gemini, and Llama2, this study explores various LLM based ontology engineering methods. The findings reveal that the proposed hybrid approach (both LLM and human involvement), namely X-HCOME, consistently excelled in class generation and F-1 score, indicating its efficiency in creating valid and comprehensive ontologies faster than humans do alone. The study underscores the potential of the combined LLMs and human intelligence to enrich PD domain knowledge and enhance expert-generated PD ontologies. In overall, the presented approach exemplifies a promising collaboration between machine capabilities and human expertise in developing ontologies for complex domains.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } This paper investigates the integration of Large Language Models (LLMs) in the engineering of a Parkinson’s Disease (PD) monitoring and alerting ontology. The focus is on the ontology engineering methodology which combines the capabilities of LLMs and human expertise to develop more robust and comprehensive domain ontologies, faster than humans do alone. Evaluating models like ChatGPT-3.5, ChatGPT4, Gemini, and Llama2, this study explores various LLM based ontology engineering methods. The findings reveal that the proposed hybrid approach (both LLM and human involvement), namely X-HCOME, consistently excelled in class generation and F-1 score, indicating its efficiency in creating valid and comprehensive ontologies faster than humans do alone. The study underscores the potential of the combined LLMs and human intelligence to enrich PD domain knowledge and enhance expert-generated PD ontologies. In overall, the presented approach exemplifies a promising collaboration between machine capabilities and human expertise in developing ontologies for complex domains. |
Tania Litaina Andreas Soularidis, Georgios Bouchouras Konstantinos Kotis Evangelia Kavakli Towards LLM-based semantic analysis of historical legal documents Workshop First International Workshop of Semantic Digital Humanities, co-located with ESWC 2024, 2024. @workshop{Litaina2024, title = {Towards LLM-based semantic analysis of historical legal documents}, author = {Tania Litaina, Andreas Soularidis, Georgios Bouchouras, Konstantinos Kotis, Evangelia Kavakli}, url = {https://www.researchgate.net/profile/Giorgos-Bouchouras/publication/383431412_Towards_LLM-based_Semantic_Analysis_of_Historical_Legal_Documents/links/66cd70cec2eaa500231aeeb6/Towards-LLM-based-Semantic-Analysis-of-Historical-Legal-Documents.pdf}, year = {2024}, date = {2024-05-26}, booktitle = {First International Workshop of Semantic Digital Humanities, co-located with ESWC 2024}, abstract = {The preservation of legal documents such as notarial ones is of vital importance as they are evidence of legal transactions between the involved entities through the years, serving as historical legal knowledge bases. The emergence of Large Language Models (LLMs) and their ability to analyze big data and generate content (much faster and relatively better than humans do alone) has created new perspectives in many fields, including law. Motivated by the significant potential of LLMs, we investigate the capabilities and limitations of using them in semantically analyzing legal documents through experimentation with two most prevalent LLMs i.e., ChatGPT-3.5 and Gemini/Bard. The goal is to emphasize automated and faster semantic analysis of documents, placing questions (prompts) concerning the type and subject of contracts, the recognition of the involved named entities and their relationship(s) e.g., landlord-tenant or family relationships. The experiments conducted with digitized contract documents that have been converted from handwritten Greek originals into plain text (LLM input) using Transkribus, an AI-powered platform for text recognition and transcription. The LLM responses were evaluated against the results obtained from a human expert, performing better in terms of precision but not in recall.}, keywords = {}, pubstate = {published}, tppubtype = {workshop} } The preservation of legal documents such as notarial ones is of vital importance as they are evidence of legal transactions between the involved entities through the years, serving as historical legal knowledge bases. The emergence of Large Language Models (LLMs) and their ability to analyze big data and generate content (much faster and relatively better than humans do alone) has created new perspectives in many fields, including law. Motivated by the significant potential of LLMs, we investigate the capabilities and limitations of using them in semantically analyzing legal documents through experimentation with two most prevalent LLMs i.e., ChatGPT-3.5 and Gemini/Bard. The goal is to emphasize automated and faster semantic analysis of documents, placing questions (prompts) concerning the type and subject of contracts, the recognition of the involved named entities and their relationship(s) e.g., landlord-tenant or family relationships. The experiments conducted with digitized contract documents that have been converted from handwritten Greek originals into plain text (LLM input) using Transkribus, an AI-powered platform for text recognition and transcription. The LLM responses were evaluated against the results obtained from a human expert, performing better in terms of precision but not in recall. |
Alexandros Karakikes Panagiotis Alexiadis, Konstantinos Kotis Bias in X (Twitter) and telegram based intelligence analysis: exploring challenges and potential mitigating roles of AI Journal Article SN Computer Science, 5 (5), pp. 574, 2024. @article{Karakikes2024, title = {Bias in X (Twitter) and telegram based intelligence analysis: exploring challenges and potential mitigating roles of AI}, author = {Alexandros Karakikes, Panagiotis Alexiadis, Konstantinos Kotis}, url = {https://link.springer.com/article/10.1007/s42979-024-02935-w}, doi = {https://doi.org/10.1007/s42979-024-02935-w}, year = {2024}, date = {2024-05-23}, journal = {SN Computer Science}, volume = {5}, number = {5}, pages = {574}, abstract = {Bias identification and mitigation in the social media ecosystem has been lately researched towards achieving a more efficient utilization of social media platforms for different stakeholders and purposes. Among these stakeholders, intelligence services worldwide, collectively called the Intelligence Community (IC), tend to use social media, supplementarily to their pre-extant disciplines, for monitoring areas of interest and identifying emerging social, political and security trends/threats. Over time, the IC has identified bias as the major impediment in information analysis, thus it has developed scientific and empirical methods for bias mitigation, in parallel to those developed by the information and communication technology (ICT) and artificial intelligence (AI) community. As it becomes apparent, it is to both communities’ interest to accurately trace bias and ideally eradicate or moderate its effects. This paper is an extension of a previously presented academic work, in which we drew systemic parallels between Intelligence Analysis (IA) and Twitter Analytics (TA), comparatively examined existing bias mitigating methodologies to identify similarities/dissimilarities, and subsequently investigated the viability of adopting and attuning methodologies from the first field to the latter. Furthermore, we proposed a novel framework for AI-augmented bias mitigation in the IC and simultaneously recommended on a theoretical level, methods and tools, already adapted by the ICT community, for efficiently supporting bias mitigation in each phase of the aforementioned framework. In the current paper, we extend our previous work by implementing the collection phase of the proposed framework on a real-world use case utilizing Telegram as a collection platform. We contribute new insights resulted from our experimentation with a tri-modal source selection approach in which human agents and Large Language Models (LLMs) are involved. The experiments were performed with data collected using one of the correspondingly suggested tools, engineering an equally represented, balanced dataset for the working case.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Bias identification and mitigation in the social media ecosystem has been lately researched towards achieving a more efficient utilization of social media platforms for different stakeholders and purposes. Among these stakeholders, intelligence services worldwide, collectively called the Intelligence Community (IC), tend to use social media, supplementarily to their pre-extant disciplines, for monitoring areas of interest and identifying emerging social, political and security trends/threats. Over time, the IC has identified bias as the major impediment in information analysis, thus it has developed scientific and empirical methods for bias mitigation, in parallel to those developed by the information and communication technology (ICT) and artificial intelligence (AI) community. As it becomes apparent, it is to both communities’ interest to accurately trace bias and ideally eradicate or moderate its effects. This paper is an extension of a previously presented academic work, in which we drew systemic parallels between Intelligence Analysis (IA) and Twitter Analytics (TA), comparatively examined existing bias mitigating methodologies to identify similarities/dissimilarities, and subsequently investigated the viability of adopting and attuning methodologies from the first field to the latter. Furthermore, we proposed a novel framework for AI-augmented bias mitigation in the IC and simultaneously recommended on a theoretical level, methods and tools, already adapted by the ICT community, for efficiently supporting bias mitigation in each phase of the aforementioned framework. In the current paper, we extend our previous work by implementing the collection phase of the proposed framework on a real-world use case utilizing Telegram as a collection platform. We contribute new insights resulted from our experimentation with a tri-modal source selection approach in which human agents and Large Language Models (LLMs) are involved. The experiments were performed with data collected using one of the correspondingly suggested tools, engineering an equally represented, balanced dataset for the working case. |
Efthymia Moraitou Yannis Christodoulou, Konstantinos Kotis George Caridakis An ontology-based framework for supporting decision-making in conservation and restoration interventions for cultural heritage Journal Article ACM Journal on Computing and Cultural Heritage, 17 (3), pp. 1-24, 2024. @article{Moraitou2024, title = {An ontology-based framework for supporting decision-making in conservation and restoration interventions for cultural heritage}, author = {Efthymia Moraitou, Yannis Christodoulou, Konstantinos Kotis, George Caridakis}, url = {https://dl.acm.org/doi/full/10.1145/3653977}, doi = {https://doi.org/10.1145/3653977}, year = {2024}, date = {2024-05-22}, journal = {ACM Journal on Computing and Cultural Heritage}, volume = {17}, number = {3}, pages = {1-24}, abstract = {Decision-making (DM) is the backbone of the Conservation and Restoration (CnR) of Cultural Heritage (CH). The demands of the DM process for information organization and management have raised issues that the CnR community attempts to solve by creating DM-support tools and systems, which, among others, exploit Semantic Web (SW) technologies. Regarding the tools and systems that focus on the DM process of selecting an intervention option (CnR-DM-I), they present benefits, as well as limitations, regarding the (1) completeness of representation of the relevant knowledge in a unified manner, (2) facilitation of recording the CnR-DM-I process per se, in terms of the problem at hand as well as the intervention parameters, requirements, and criteria, and (3) recommendation and further exploration of CnR intervention options in a systematic manner. This work proposes an ontology-based framework as a means to overcome those limitations. The proposed framework (DS-CnRI) sets at its core a formal ontology which provides the necessary entities to represent expert knowledge related to CnR-DM-I. The ontology also includes rules which provide useful inferences to assist the CnR-DM-I process. The proposed framework has been deployed and evaluated in collaboration with conservators. Initial evaluation results show that the framework assists conservators in CnR-DM-I to detect and select the most suitable intervention options, to better understand the limitations of different options, and to document the process of reaching their decision.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Decision-making (DM) is the backbone of the Conservation and Restoration (CnR) of Cultural Heritage (CH). The demands of the DM process for information organization and management have raised issues that the CnR community attempts to solve by creating DM-support tools and systems, which, among others, exploit Semantic Web (SW) technologies. Regarding the tools and systems that focus on the DM process of selecting an intervention option (CnR-DM-I), they present benefits, as well as limitations, regarding the (1) completeness of representation of the relevant knowledge in a unified manner, (2) facilitation of recording the CnR-DM-I process per se, in terms of the problem at hand as well as the intervention parameters, requirements, and criteria, and (3) recommendation and further exploration of CnR intervention options in a systematic manner. This work proposes an ontology-based framework as a means to overcome those limitations. The proposed framework (DS-CnRI) sets at its core a formal ontology which provides the necessary entities to represent expert knowledge related to CnR-DM-I. The ontology also includes rules which provide useful inferences to assist the CnR-DM-I process. The proposed framework has been deployed and evaluated in collaboration with conservators. Initial evaluation results show that the framework assists conservators in CnR-DM-I to detect and select the most suitable intervention options, to better understand the limitations of different options, and to document the process of reaching their decision. |
Piyabhum Chaysri Christos Spatharis, Kostas Vlachos Konstantinos Blekas Design and implementation of a low-cost intelligent unmanned surface vehicle Journal Article Sensors, 24 (10), pp. 3254, 2024. @article{Chaysri2024, title = {Design and implementation of a low-cost intelligent unmanned surface vehicle}, author = {Piyabhum Chaysri, Christos Spatharis, Kostas Vlachos, Konstantinos Blekas}, url = {https://www.mdpi.com/1424-8220/24/10/3254}, doi = {https://doi.org/10.3390/s24103254}, year = {2024}, date = {2024-05-20}, journal = {Sensors}, volume = {24}, number = {10}, pages = {3254}, abstract = {This article describes the design and construction journey of a self-developed unmanned surface vehicle (USV). In order to increase the accessibility and lower the barrier of entry we propose a low-cost (under EUR 1000) approach to the vessel construction with great adaptability and customizability. This design prioritizes minimal power consumption as a key objective. It focuses on elucidating the intricacies of both the design and assembly processes involved in creating an economical USV. Utilizing easily accessible components, the boat outlined in this study has been already participated in the 1st Aegean Ro-boat Race 2023 competition and is tailored for entry into similar robotic competitions. Its primary functionalities encompass autonomous sea navigation coupled with sophisticated collision avoidance capabilities. Finally, we studied reinforcement learning strategies for constructing a robust intelligent controller for the task of USV navigation under disturbances and we show some preliminary simulation results we have obtained.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This article describes the design and construction journey of a self-developed unmanned surface vehicle (USV). In order to increase the accessibility and lower the barrier of entry we propose a low-cost (under EUR 1000) approach to the vessel construction with great adaptability and customizability. This design prioritizes minimal power consumption as a key objective. It focuses on elucidating the intricacies of both the design and assembly processes involved in creating an economical USV. Utilizing easily accessible components, the boat outlined in this study has been already participated in the 1st Aegean Ro-boat Race 2023 competition and is tailored for entry into similar robotic competitions. Its primary functionalities encompass autonomous sea navigation coupled with sophisticated collision avoidance capabilities. Finally, we studied reinforcement learning strategies for constructing a robust intelligent controller for the task of USV navigation under disturbances and we show some preliminary simulation results we have obtained. |
Sotirios Bentos Eleftherios Bailis, Fotini Oikonomou Stamatis Spirou Emmanouil Mavrikos Stamatis Chatzistamatis Konstantinos Kotis George Tsekouras E 2024 4th International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET), 2024, ISBN: 979-8-3503-0950-8. @conference{Bentos2024, title = {Evaluation of fairness in machine learning-based recidivism predictions: The case of Greek female prison system}, author = {Sotirios Bentos, Eleftherios Bailis, Fotini Oikonomou, Stamatis Spirou, Emmanouil Mavrikos, Stamatis Chatzistamatis, Konstantinos Kotis, George E Tsekouras}, url = {https://ieeexplore.ieee.org/abstract/document/10548202}, doi = {https://doi.org/10.1109/IRASET60544.2024.10548202}, isbn = {979-8-3503-0950-8}, year = {2024}, date = {2024-05-16}, booktitle = {2024 4th International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET)}, pages = {1-8}, abstract = {Recidivism refers to a person’s relapse into criminal behavior, often after receiving some form of punishment or undergoing intervention for a previous crime. Machine learning (ML) algorithms are commonly used for quantitatively predicting recidivism by assessing a criminal defendant’s likelihood of committing a crime thus, guiding decisions and imposing choices for criminal justice officers in managing the criminal population. Beyond the prediction adequacy of these algorithms, an important issue is whether they are capable of making fair decisions. It has been stated that attributes such as gender, race, age, ethnicity, and unemployment appear to affect the fair decision-making of ML systems upon recidivism. In this paper, we study the recidivism predictions obtained by several supervised ML algorithms over a dataset that has been extracted from a Greek female prison data record. The main points addressed by the current contribution concern the study of the resulting recidivism predictions from the perspective of fairness assessment that is related to certain data attributes such as age at exiting the first imprisonment, and employment status at the moment of the first imprisonment. To accomplish that task, several criteria are applied to analyze the ML-based predictions in terms of statistical analysis.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } Recidivism refers to a person’s relapse into criminal behavior, often after receiving some form of punishment or undergoing intervention for a previous crime. Machine learning (ML) algorithms are commonly used for quantitatively predicting recidivism by assessing a criminal defendant’s likelihood of committing a crime thus, guiding decisions and imposing choices for criminal justice officers in managing the criminal population. Beyond the prediction adequacy of these algorithms, an important issue is whether they are capable of making fair decisions. It has been stated that attributes such as gender, race, age, ethnicity, and unemployment appear to affect the fair decision-making of ML systems upon recidivism. In this paper, we study the recidivism predictions obtained by several supervised ML algorithms over a dataset that has been extracted from a Greek female prison data record. The main points addressed by the current contribution concern the study of the resulting recidivism predictions from the perspective of fairness assessment that is related to certain data attributes such as age at exiting the first imprisonment, and employment status at the moment of the first imprisonment. To accomplish that task, several criteria are applied to analyze the ML-based predictions in terms of statistical analysis. |
Andreas Kontogiannis Vasilis Pollatos, Sotiris Kanellopoulos Panayotis Mertikopoulos Aris Pagourtzis Ioannis Panageas The computational complexity of finding second-order stationary points Conference Forty-first International Conference on Machine Learning, 2024. @conference{Kontogiannis*2024, title = {The computational complexity of finding second-order stationary points}, author = {Andreas Kontogiannis, Vasilis Pollatos, Sotiris Kanellopoulos, Panayotis Mertikopoulos, Aris Pagourtzis, Ioannis Panageas}, url = {https://openreview.net/forum?id=t8WDBcegae}, year = {2024}, date = {2024-05-02}, booktitle = {Forty-first International Conference on Machine Learning}, abstract = {Non-convex minimization problems are universally considered hard, and even guaranteeing that a computed solution is locally minimizing is known to be NP-hard. In this general context, our paper focuses on the problem of finding stationary points that satisfy an approximate second-order optimality condition, which serves to exclude strict saddles and other non-minimizing stationary points. Our main result is that the problem of finding approximate second-order stationary points (SOSPs) is PLS-complete, i.e., of the same complexity as the problem of finding first-order stationary points (FOSPs), thus resolving an open question in the field. In particular, our results imply that, under the widely believed complexity conjecture that PLS FNP, finding approximate SOSPs in unconstrained domains is *easier* than in constrained domains, which is known to be NP-hard. This comes in stark contrast with earlier results which implied that, unless PLS = CLS, finding approximate FOSPs in unconstrained domains is *harder* than in constrained domains.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } Non-convex minimization problems are universally considered hard, and even guaranteeing that a computed solution is locally minimizing is known to be NP-hard. In this general context, our paper focuses on the problem of finding stationary points that satisfy an approximate second-order optimality condition, which serves to exclude strict saddles and other non-minimizing stationary points. Our main result is that the problem of finding approximate second-order stationary points (SOSPs) is PLS-complete, i.e., of the same complexity as the problem of finding first-order stationary points (FOSPs), thus resolving an open question in the field. In particular, our results imply that, under the widely believed complexity conjecture that PLS FNP, finding approximate SOSPs in unconstrained domains is *easier* than in constrained domains, which is known to be NP-hard. This comes in stark contrast with earlier results which implied that, unless PLS = CLS, finding approximate FOSPs in unconstrained domains is *harder* than in constrained domains. |
Konstantinos Bougiatiotis Andreas Sideras, Elias Zavitsanos Georgios Paliouras Proceedings of the Joint Workshop of the 7th Financial Technology and Natural Language Processing, the 5th Knowledge Discovery from Unstructured Data in Financial Services, and the 4th Workshop on Economics and Natural Language Processing, 2024. @conference{Bougiatiotis2024, title = {Dice@ ml-esg-3: Esg impact level and duration inference using llms for augmentation and contrastive learning}, author = {Konstantinos Bougiatiotis, Andreas Sideras, Elias Zavitsanos, Georgios Paliouras}, url = {https://aclanthology.org/2024.finnlp-1.24/ https://aclanthology.org/2024.finnlp-1.24.pdf}, year = {2024}, date = {2024-05-01}, booktitle = {Proceedings of the Joint Workshop of the 7th Financial Technology and Natural Language Processing, the 5th Knowledge Discovery from Unstructured Data in Financial Services, and the 4th Workshop on Economics and Natural Language Processing}, pages = {234-243}, abstract = {We present the submission of team DICE for ML-ESG-3, the 3rd Shared Task on Multilingual ESG impact duration inference in the context of the joint FinNLP-KDF workshop series. The task provides news articles and seeks to determine the impact and duration of an event in the news article may have on a company. We experiment with various baselines and discuss the results of our best-performing submissions based on contrastive pre-training and a stacked model based on the bag-of-words assumption and sentence embeddings. We also explored the label correlations among events stemming from the same news article and the correlations between impact level and impact length. Our analysis shows that even simple classifiers trained in this task can achieve comparable performance with more complex models, under certain conditions.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } We present the submission of team DICE for ML-ESG-3, the 3rd Shared Task on Multilingual ESG impact duration inference in the context of the joint FinNLP-KDF workshop series. The task provides news articles and seeks to determine the impact and duration of an event in the news article may have on a company. We experiment with various baselines and discuss the results of our best-performing submissions based on contrastive pre-training and a stacked model based on the bag-of-words assumption and sentence embeddings. We also explored the label correlations among events stemming from the same news article and the correlations between impact level and impact length. Our analysis shows that even simple classifiers trained in this task can achieve comparable performance with more complex models, under certain conditions. |
Sotiris Angelis Efthymia Moraitou, George Caridakis Konstantinos Kotis CHEKG: a collaborative and hybrid methodology for engineering modular and fair domain-specific knowledge graphs Journal Article Knowledge and Information Systems, 66 , pp. 4899–4925, 2024. @article{Angelis2024, title = {CHEKG: a collaborative and hybrid methodology for engineering modular and fair domain-specific knowledge graphs}, author = {Sotiris Angelis, Efthymia Moraitou, George Caridakis, Konstantinos Kotis}, url = {https://link.springer.com/content/pdf/10.1007/s10115-024-02110-w.pdf}, doi = {https://doi.org/10.1007/s10115-024-02110-w}, year = {2024}, date = {2024-04-20}, journal = {Knowledge and Information Systems}, volume = {66}, pages = {4899–4925}, abstract = {Ontologies constitute the semantic model of Knowledge Graphs (KGs). This structural association indicates the potential existence of methodological analogies in the development of ontologies and KGs. The deployment of fully and well-defined methodologies for KG development based on existing ontology engineering methodologies (OEMs) has been suggested and efficiently applied. However, most of the modern/recent OEMs may not include tasks that (i) empower knowledge workers and domain experts to closely collaborate with ontology engineers and KG specialists for the development and maintenance of KGs, (ii) satisfy special requirements of KG development, such as (a) ensuring modularity and agility of KGs, (b) assessing and mitigating bias at schema and data levels. Toward this aim, the paper presents a methodology for the Collaborative and Hybrid Engineering of Knowledge Graphs (CHEKG), which constitutes a hybrid (schema-centric/top-down and data-driven/bottom-up), collaborative, agile, and iterative approach for developing modular and fair domain-specific KGs. CHEKG contributes to all phases of the KG engineering lifecycle: from the specification of a KG to its exploitation, evaluation, and refinement. The CHEKG methodology is based on the main phases of the extended Human-Centered Collaborative Ontology Engineering Methodology (ext-HCOME), while it adjusts and expands the individual processes and tasks of each phase according to the specialized requirements of KG development. Apart from the presentation of the methodology per se, the paper presents recent work regarding the deployment and evaluation of the CHEKG methodology for the engineering of semantic trajectories as KGs generated from unmanned aerial vehicles (UAVs) data during real cultural heritage documentation scenarios.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Ontologies constitute the semantic model of Knowledge Graphs (KGs). This structural association indicates the potential existence of methodological analogies in the development of ontologies and KGs. The deployment of fully and well-defined methodologies for KG development based on existing ontology engineering methodologies (OEMs) has been suggested and efficiently applied. However, most of the modern/recent OEMs may not include tasks that (i) empower knowledge workers and domain experts to closely collaborate with ontology engineers and KG specialists for the development and maintenance of KGs, (ii) satisfy special requirements of KG development, such as (a) ensuring modularity and agility of KGs, (b) assessing and mitigating bias at schema and data levels. Toward this aim, the paper presents a methodology for the Collaborative and Hybrid Engineering of Knowledge Graphs (CHEKG), which constitutes a hybrid (schema-centric/top-down and data-driven/bottom-up), collaborative, agile, and iterative approach for developing modular and fair domain-specific KGs. CHEKG contributes to all phases of the KG engineering lifecycle: from the specification of a KG to its exploitation, evaluation, and refinement. The CHEKG methodology is based on the main phases of the extended Human-Centered Collaborative Ontology Engineering Methodology (ext-HCOME), while it adjusts and expands the individual processes and tasks of each phase according to the specialized requirements of KG development. Apart from the presentation of the methodology per se, the paper presents recent work regarding the deployment and evaluation of the CHEKG methodology for the engineering of semantic trajectories as KGs generated from unmanned aerial vehicles (UAVs) data during real cultural heritage documentation scenarios. |
Batsis, Georgios Machine learning for children’s music emotion recognition Masters Thesis Department of Digital Systems, School of Information Technologies and Communication, University of Piraeus, 2024. @mastersthesis{Batsis2024, title = {Machine learning for children’s music emotion recognition}, author = {Georgios Batsis}, url = {https://dione.lib.unipi.gr/xmlui/handle/unipi/16460}, doi = {http://dx.doi.org/10.26267/unipi_dione/3882}, year = {2024}, date = {2024-04-01}, school = {Department of Digital Systems, School of Information Technologies and Communication, University of Piraeus}, abstract = {This work focuses on the application of Machine Learning techniques for Music Emotion Recognition, particularly focusing on children’s music. The first step was to create a specialized dataset for children’s music, which includes songs of varied emotions and cultural backgrounds, annotated by experts in child psychology, education, and Machine Learning Engineers. A Support Vector Machine was employed as a baseline model for the prediction task, processing a range of handcrafted audio features. Concerning more advanced models, Convolutional Neural Networks and a Dual-Stream architecture model, integrating both Convolutional and attention-based Long Short-Term Memory networks were evaluated. This approach offers a comprehensive analysis of children’s music by examining both spectrograms and music transcription sequences. Models were evaluated using the Probabilistic Emotion Alignment to compare model posteriors with the probability distribution of expert annotations. Moreover, models evaluated using the established Machine Learning metrics, indicating that different modalities are able to enhance the predictive capacity for emotion recognition.}, keywords = {}, pubstate = {published}, tppubtype = {mastersthesis} } This work focuses on the application of Machine Learning techniques for Music Emotion Recognition, particularly focusing on children’s music. The first step was to create a specialized dataset for children’s music, which includes songs of varied emotions and cultural backgrounds, annotated by experts in child psychology, education, and Machine Learning Engineers. A Support Vector Machine was employed as a baseline model for the prediction task, processing a range of handcrafted audio features. Concerning more advanced models, Convolutional Neural Networks and a Dual-Stream architecture model, integrating both Convolutional and attention-based Long Short-Term Memory networks were evaluated. This approach offers a comprehensive analysis of children’s music by examining both spectrograms and music transcription sequences. Models were evaluated using the Probabilistic Emotion Alignment to compare model posteriors with the probability distribution of expert annotations. Moreover, models evaluated using the established Machine Learning metrics, indicating that different modalities are able to enhance the predictive capacity for emotion recognition. |
N. Zafeiropoulos P. Bitilis, Tsekouras Kotis G E K Evaluating Ontology-Based PD Monitoring and Alerting in Personal Health Knowledge Graphs and Graph Neural Networks Journal Article Information, 15 (2), pp. 100, 2024. @article{Zafeiropoulos2024, title = {Evaluating Ontology-Based PD Monitoring and Alerting in Personal Health Knowledge Graphs and Graph Neural Networks}, author = {N. Zafeiropoulos, P. Bitilis, G.E. Tsekouras, K. Kotis}, url = {https://www.mdpi.com/2078-2489/15/2/100/pdf?version=1707998600}, doi = {https://doi.org/10.3390/info15020100}, year = {2024}, date = {2024-02-08}, journal = {Information}, volume = {15}, number = {2}, pages = {100}, abstract = {In the realm of Parkinson’s Disease (PD) research, the integration of wearable sensor data with personal health records (PHR) has emerged as a pivotal avenue for patient alerting and monitoring. This study delves into the complex domain of PD patient care, with a specific emphasis on harnessing the potential of wearable sensors to capture, represent and semantically analyze crucial movement data and knowledge. The primary objective is to enhance the assessment of PD patients by establishing a robust foundation for personalized health insights through the development of Personal Health Knowledge Graphs (PHKGs) and the employment of personal health Graph Neural Networks (PHGNNs) that utilize PHKGs. The objective is to formalize the representation of related integrated data, unified sensor and PHR data in higher levels of abstraction, i.e., in a PHKG, to facilitate interoperability and support rule-based high-level event recognition such as patient’s missing dose or falling. This paper, extending our previous related work, presents the Wear4PDmove ontology in detail and evaluates the ontology within the development of an experimental PHKG. Furthermore, this paper focuses on the integration and evaluation of PHKG within the implementation of a Graph Neural Network (GNN). This work emphasizes the importance of integrating PD-related data for monitoring and alerting patients with appropriate notifications. These notifications offer health experts precise and timely information for the continuous evaluation of personal health-related events, ultimately contributing to enhanced patient care and well-informed medical decision-making. Finally, the paper concludes by proposing a novel approach for integrating personal health KGs and GNNs for PD monitoring and alerting solutions.}, keywords = {}, pubstate = {published}, tppubtype = {article} } In the realm of Parkinson’s Disease (PD) research, the integration of wearable sensor data with personal health records (PHR) has emerged as a pivotal avenue for patient alerting and monitoring. This study delves into the complex domain of PD patient care, with a specific emphasis on harnessing the potential of wearable sensors to capture, represent and semantically analyze crucial movement data and knowledge. The primary objective is to enhance the assessment of PD patients by establishing a robust foundation for personalized health insights through the development of Personal Health Knowledge Graphs (PHKGs) and the employment of personal health Graph Neural Networks (PHGNNs) that utilize PHKGs. The objective is to formalize the representation of related integrated data, unified sensor and PHR data in higher levels of abstraction, i.e., in a PHKG, to facilitate interoperability and support rule-based high-level event recognition such as patient’s missing dose or falling. This paper, extending our previous related work, presents the Wear4PDmove ontology in detail and evaluates the ontology within the development of an experimental PHKG. Furthermore, this paper focuses on the integration and evaluation of PHKG within the implementation of a Graph Neural Network (GNN). This work emphasizes the importance of integrating PD-related data for monitoring and alerting patients with appropriate notifications. These notifications offer health experts precise and timely information for the continuous evaluation of personal health-related events, ultimately contributing to enhanced patient care and well-informed medical decision-making. Finally, the paper concludes by proposing a novel approach for integrating personal health KGs and GNNs for PD monitoring and alerting solutions. |
Search with keywords
| 1. | Andreas Kontogiannis Vasilis Pollatos, Panayotis Mertikopoulos Ioannis Panageas : Efficient swap regret minimization in combinatorial bandits. Twenty-Ninth Annual Conference on Artificial Intelligence and Statistics (AISTATS 2026), 2026. (Type: Conference | Abstract | Links | BibTeX) @conference{Kontogiannis2026, title = {Efficient swap regret minimization in combinatorial bandits}, author = {Andreas Kontogiannis, Vasilis Pollatos, Panayotis Mertikopoulos, Ioannis Panageas}, url = {https://arxiv.org/pdf/2602.02087}, year = {2026}, date = {2026-05-02}, booktitle = {Twenty-Ninth Annual Conference on Artificial Intelligence and Statistics (AISTATS 2026)}, abstract = {This paper addresses the problem of designing efficient no-swap regret algorithms for combinatorial bandits, where the number of actions N is exponentially large in the dimensionality of the problem. In this setting, designing efficient no-swap regret translates to sublinear — in horizon T — swap regret with polylogarithmic dependence on N. In contrast to the weaker notion of external regret minimization – a problem which is fairly well understood in the literature – achieving no-swap regret with a polylogarithmic dependence on N has remained elusive in combinatorial bandits. Our paper resolves this challenge, by introducing a no-swap-regret learning algorithm with regret that scales polylogarithmically in N and is tight for the class of combinatorial bandits. To ground our results, we also demonstrate how to implement the proposed algorithm efficiently — that is, with a per-iteration complexity that also scales polylogarithmically in N — across a wide range of well-studied applications.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } This paper addresses the problem of designing efficient no-swap regret algorithms for combinatorial bandits, where the number of actions N is exponentially large in the dimensionality of the problem. In this setting, designing efficient no-swap regret translates to sublinear — in horizon T — swap regret with polylogarithmic dependence on N. In contrast to the weaker notion of external regret minimization – a problem which is fairly well understood in the literature – achieving no-swap regret with a polylogarithmic dependence on N has remained elusive in combinatorial bandits. Our paper resolves this challenge, by introducing a no-swap-regret learning algorithm with regret that scales polylogarithmically in N and is tight for the class of combinatorial bandits. To ground our results, we also demonstrate how to implement the proposed algorithm efficiently — that is, with a per-iteration complexity that also scales polylogarithmically in N — across a wide range of well-studied applications. |
| 2. | Andreas Soularidis Dimitrios Doumanas, Konstantinos Kotis George Vouros A: Automating agentic collaborative ontology engineering with role-playing simulation of LLM-powered agents and RAG technology. In: The Knowledge Engineering Review, 40 , pp. e10, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Soularidis2025, title = {Automating agentic collaborative ontology engineering with role-playing simulation of LLM-powered agents and RAG technology}, author = {Andreas Soularidis, Dimitrios Doumanas, Konstantinos Kotis, George A Vouros}, doi = {https://doi.org/10.1017/S026988892510009X}, year = {2025}, date = {2025-12-19}, journal = {The Knowledge Engineering Review}, volume = {40}, pages = {e10}, abstract = {Motivated by the astonishing capabilities of large language models (LLMs) in text-generation, reasoning, and simulation of complex human behaviors, in this paper, we propose a novel multi-component LLM-based framework, namely LLM4ACOE, that fully automates the collaborative ontology engineering (COE) process using role-playing simulation of LLM agents and retrieval augmented generation (RAG) technology. The proposed solution enhances the LLM-powered role-playing simulation with RAG ‘feeding’ the LLM with three different types of external knowledge. This knowledge corresponds to the knowledge required by each of the COE roles (agents), using a component-based framework, as follows: (a) domain-specific data-centric documents, (b) OWL documentation, and (c) ReAct guidelines. The aforementioned components are evaluated in combination, with the aim of investigating their impact on the quality of generated ontologies. The aim of this work is twofold, (a) to identify the capacity of LLM-based agents to generate acceptable (by human-experts) ontologies through agentic collaborative ontology engineering (ACOE) role-playing simulation, at specific levels of acceptance (accuracy, validity, and expressiveness of ontologies) without human intervention and (b) to investigate whether and/or to what extent the selected RAG components affect the quality of the generated ontologies. The evaluation of this novel approach is performed using ChatGPT-o in the domain of search and rescue (SAR) missions. To assess the generated ontologies, quantitative and qualitative measures are employed, focusing on coverage, expressiveness, structure, and human involvement.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Motivated by the astonishing capabilities of large language models (LLMs) in text-generation, reasoning, and simulation of complex human behaviors, in this paper, we propose a novel multi-component LLM-based framework, namely LLM4ACOE, that fully automates the collaborative ontology engineering (COE) process using role-playing simulation of LLM agents and retrieval augmented generation (RAG) technology. The proposed solution enhances the LLM-powered role-playing simulation with RAG ‘feeding’ the LLM with three different types of external knowledge. This knowledge corresponds to the knowledge required by each of the COE roles (agents), using a component-based framework, as follows: (a) domain-specific data-centric documents, (b) OWL documentation, and (c) ReAct guidelines. The aforementioned components are evaluated in combination, with the aim of investigating their impact on the quality of generated ontologies. The aim of this work is twofold, (a) to identify the capacity of LLM-based agents to generate acceptable (by human-experts) ontologies through agentic collaborative ontology engineering (ACOE) role-playing simulation, at specific levels of acceptance (accuracy, validity, and expressiveness of ontologies) without human intervention and (b) to investigate whether and/or to what extent the selected RAG components affect the quality of the generated ontologies. The evaluation of this novel approach is performed using ChatGPT-o in the domain of search and rescue (SAR) missions. To assess the generated ontologies, quantitative and qualitative measures are employed, focusing on coverage, expressiveness, structure, and human involvement. |
| 3. | Apostolos Glenis, George Vouros : Scalable Univariate and Multivariate Time-Series Classifiers with Deep Learning Methods Exploiting Symbolic Representations. In: Computers, 14 (12), pp. 563, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Glenis2025, title = {Scalable Univariate and Multivariate Time-Series Classifiers with Deep Learning Methods Exploiting Symbolic Representations}, author = {Apostolos Glenis, George Vouros}, url = {https://www.mdpi.com/2073-431X/14/12/563}, doi = {https://doi.org/10.3390/computers14120563}, year = {2025}, date = {2025-12-17}, journal = {Computers}, volume = {14}, number = {12}, pages = {563}, abstract = {Time-series classification (TSC) is an important task across sciences. Symbolic representations (especially SFA) are very effective at combating noise. In this paper, we employ symbolic representations to create state-of-the-art time-series classifiers, with the aim to advance scalability without sacrificing accuracy. First, we create a graph representation of the time series based on SFA words. We use this representation together with graph kernels and an SVM classifier to create a scalable time-series classifier. Next, we use the graph representation together with a Graph Convolutional Neural Network to test how it fares against state-of-the-art time-series classifiers. Additionally, we devised deep neural networks exploiting the SFA representation, inspired by the text classification domain, to study how they fare against state-of-the-art classifiers. The proposed deep learning classifiers have been adapted and evaluated for the multivariate time-series case and also against state-of-the-art time-series classification algorithms based on symbolic representations.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Time-series classification (TSC) is an important task across sciences. Symbolic representations (especially SFA) are very effective at combating noise. In this paper, we employ symbolic representations to create state-of-the-art time-series classifiers, with the aim to advance scalability without sacrificing accuracy. First, we create a graph representation of the time series based on SFA words. We use this representation together with graph kernels and an SVM classifier to create a scalable time-series classifier. Next, we use the graph representation together with a Graph Convolutional Neural Network to test how it fares against state-of-the-art time-series classifiers. Additionally, we devised deep neural networks exploiting the SFA representation, inspired by the text classification domain, to study how they fare against state-of-the-art classifiers. The proposed deep learning classifiers have been adapted and evaluated for the multivariate time-series case and also against state-of-the-art time-series classification algorithms based on symbolic representations. |
| 4. | Georgios Bouchouras Dimitrios Doumanas, Andreas Soularidis Konstantinos Kotis George Vouros A: Leveraging LLMs for Collaborative Ontology Engineering in Parkinson Disease Monitoring and Alerting. In: arXiv, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Bouchouras2025, title = {Leveraging LLMs for Collaborative Ontology Engineering in Parkinson Disease Monitoring and Alerting}, author = {Georgios Bouchouras, Dimitrios Doumanas, Andreas Soularidis, Konstantinos Kotis, George A Vouros}, url = {https://arxiv.org/pdf/2512.14288}, doi = {https://doi.org/10.48550/arXiv.2512.14288}, year = {2025}, date = {2025-12-16}, journal = {arXiv}, abstract = {This paper explores the integration of Large Language Models (LLMs) in the engineering of a Parkinson’s Disease (PD) monitoring and alerting ontology through four key methodologies: One Shot (OS) prompt techniques, Chain of Thought (CoT) prompts, X-HCOME, and SimX-HCOME+. The primary objective is to determine whether LLMs alone can create comprehensive ontologies and, if not, whether human-LLM collaboration can achieve this goal. Consequently, the paper assesses the effectiveness of LLMs in automated ontology development and the enhancement achieved through human-LLM collaboration. Initial ontology generation was performed using One Shot (OS) and Chain of Thought (CoT) prompts, demonstrating the capability of LLMs to autonomously construct ontologies for PD monitoring and alerting. However, these outputs were not comprehensive and required substantial human refinement to enhance their completeness and accuracy. X-HCOME, a hybrid ontology engineering approach that combines human expertise with LLM capabilities, showed significant improvements in ontology comprehensiveness. This methodology resulted in ontologies that are very similar to those constructed by experts. Further experimentation with SimX-HCOME+, another hybrid methodology emphasizing continuous human supervision and iterative refinement, highlighted the importance of ongoing human involvement. This approach led to the creation of more comprehensive and accurate ontologies. Overall, the paper underscores the potential of human-LLM collaboration in advancing ontology engineering, particularly in complex domains like PD. The results suggest promising directions for future research, including the development of specialized GPT models for ontology construction.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This paper explores the integration of Large Language Models (LLMs) in the engineering of a Parkinson’s Disease (PD) monitoring and alerting ontology through four key methodologies: One Shot (OS) prompt techniques, Chain of Thought (CoT) prompts, X-HCOME, and SimX-HCOME+. The primary objective is to determine whether LLMs alone can create comprehensive ontologies and, if not, whether human-LLM collaboration can achieve this goal. Consequently, the paper assesses the effectiveness of LLMs in automated ontology development and the enhancement achieved through human-LLM collaboration. Initial ontology generation was performed using One Shot (OS) and Chain of Thought (CoT) prompts, demonstrating the capability of LLMs to autonomously construct ontologies for PD monitoring and alerting. However, these outputs were not comprehensive and required substantial human refinement to enhance their completeness and accuracy. X-HCOME, a hybrid ontology engineering approach that combines human expertise with LLM capabilities, showed significant improvements in ontology comprehensiveness. This methodology resulted in ontologies that are very similar to those constructed by experts. Further experimentation with SimX-HCOME+, another hybrid methodology emphasizing continuous human supervision and iterative refinement, highlighted the importance of ongoing human involvement. This approach led to the creation of more comprehensive and accurate ontologies. Overall, the paper underscores the potential of human-LLM collaboration in advancing ontology engineering, particularly in complex domains like PD. The results suggest promising directions for future research, including the development of specialized GPT models for ontology construction. |
| 5. | Andreas Sideras Konstantinos Bougiatiotis, Elias Zavitsanos Georgios Paliouras George Vouros : A Multimodal Alignment-Based Anomaly Detection Method for Bankruptcy Prediction. Proceedings of the 6th ACM International Conference on AI in Finance, 2025, ISBN: 9798400722202. (Type: Conference | Abstract | Links | BibTeX) @conference{Sideras2025, title = {A Multimodal Alignment-Based Anomaly Detection Method for Bankruptcy Prediction}, author = {Andreas Sideras, Konstantinos Bougiatiotis, Elias Zavitsanos, Georgios Paliouras, George Vouros}, url = {https://dl.acm.org/doi/full/10.1145/3768292.3770380}, doi = {https://doi.org/10.1145/3768292.3770380}, isbn = {9798400722202}, year = {2025}, date = {2025-11-15}, booktitle = {Proceedings of the 6th ACM International Conference on AI in Finance}, pages = {53-61}, abstract = {We present a novel anomaly detection method for next-year bankruptcy prediction, utilizing a combination of financial figures and textual content from annual reports. Our approach, MABAD, learns a shared representation space where non-bankrupt firms share position and orientation. Samples that deviate from this pattern are assigned a higher anomaly score. The proposed method is tailored for highly imbalanced scenarios and is robust to heterogeneous, incomplete, and potentially contradictory inputs. We demonstrate that MABAD consistently outperforms a range of strong baselines, and we also curate and release a new publicly available multisource dataset to foster further research in the domain.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } We present a novel anomaly detection method for next-year bankruptcy prediction, utilizing a combination of financial figures and textual content from annual reports. Our approach, MABAD, learns a shared representation space where non-bankrupt firms share position and orientation. Samples that deviate from this pattern are assigned a higher anomaly score. The proposed method is tailored for highly imbalanced scenarios and is robust to heterogeneous, incomplete, and potentially contradictory inputs. We demonstrate that MABAD consistently outperforms a range of strong baselines, and we also curate and release a new publicly available multisource dataset to foster further research in the domain. |
| 6. | Elias Zavitsanos Konstantinos Bougiatiotis, Andreas Sideras Georgios Paliouras : Positive-Unlabeled Learning for Financial Misstatement Detection under Realistic Constraints. ICAIF ’25: Proceedings of the 6th ACM International Conference on AI in Finance, 2025, ISBN: 9798400722202. (Type: Conference | Abstract | Links | BibTeX) @conference{Zavitsanos2025, title = {Positive-Unlabeled Learning for Financial Misstatement Detection under Realistic Constraints}, author = {Elias Zavitsanos, Konstantinos Bougiatiotis, Andreas Sideras, Georgios Paliouras}, url = {https://dl.acm.org/doi/full/10.1145/3768292.3770366 https://dl.acm.org/doi/epdf/10.1145/3768292.3770366}, doi = {https://doi.org/10.1145/3768292.3770366}, isbn = {9798400722202}, year = {2025}, date = {2025-11-15}, booktitle = {ICAIF ’25: Proceedings of the 6th ACM International Conference on AI in Finance}, pages = {864-872}, abstract = {Detecting financial misstatements is critical for market integrity but remains challenging due to class imbalance, delayed discovery, and limited labeled data. We propose a novel Positive-Unlabeled (PU) learning framework that models the detection task under realistic constraints, where only a small subset of misstatements is known at training time. Our approach integrates unlabeled data into training, preserves temporal structure, and accounts for extreme imbalance. We construct and release a benchmark dataset reflecting these characteristics and evaluate several PU learning methods against recent baselines. Results show that PU-based models consistently outperform supervised approaches, highlighting their suitability for real-world misstatement detection.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } Detecting financial misstatements is critical for market integrity but remains challenging due to class imbalance, delayed discovery, and limited labeled data. We propose a novel Positive-Unlabeled (PU) learning framework that models the detection task under realistic constraints, where only a small subset of misstatements is known at training time. Our approach integrates unlabeled data into training, preserves temporal structure, and accounts for extreme imbalance. We construct and release a benchmark dataset reflecting these characteristics and evaluate several PU learning methods against recent baselines. Results show that PU-based models consistently outperform supervised approaches, highlighting their suitability for real-world misstatement detection. |
| 7. | Theodore Tranos Nikolaos Fesakis, Thomas Vasileiou Sotirios Christopoulos Georgio Loukos Maria Koutsoupidou : AI-Based Energy Forecasting at Different Distribution Grid Levels to Support Baseline Definition and DSO Participation in LFMs. 2025 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe), IEEE, 2025, ISBN: 979-8-3315-2503-3. (Type: Conference | Abstract | Links | BibTeX) @conference{Tranos2025, title = {AI-Based Energy Forecasting at Different Distribution Grid Levels to Support Baseline Definition and DSO Participation in LFMs}, author = {Theodore Tranos, Nikolaos Fesakis, Thomas Vasileiou, Sotirios Christopoulos, Georgio Loukos, Maria Koutsoupidou}, url = {https://ieeexplore.ieee.org/abstract/document/11305676}, doi = {https://doi.org/10.1109/ISGTEurope64741.2025.11305676}, isbn = {979-8-3315-2503-3}, year = {2025}, date = {2025-10-20}, booktitle = {2025 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe)}, pages = {1-5}, publisher = {IEEE}, abstract = {A crucial aspect of Local Flexibility Markets (LFMs) is the definition of a baseline for energy production and demand forecasting, which serves as a reference for validating and compensating flexibility services. In this study, we explore the application of machine learning techniques, specifically Long Short-Term Memory (LSTM) networks, to establish accurate baselines for consumers and producers connected to the LV grid. The LSTM models leverage real historical demand and generation data from DSO smart meters in Mesogeia, Greece, combined with weather variables such as temperature and cloud coverage, to enhance forecasting accuracy. Our goal is to evaluate forecasting accuracy at the individual participant level and compare it with the accuracy obtained from forecasting on aggregated consumption/production data within a specific grid segment or using data from the secondary substation to which the participants are connected.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } A crucial aspect of Local Flexibility Markets (LFMs) is the definition of a baseline for energy production and demand forecasting, which serves as a reference for validating and compensating flexibility services. In this study, we explore the application of machine learning techniques, specifically Long Short-Term Memory (LSTM) networks, to establish accurate baselines for consumers and producers connected to the LV grid. The LSTM models leverage real historical demand and generation data from DSO smart meters in Mesogeia, Greece, combined with weather variables such as temperature and cloud coverage, to enhance forecasting accuracy. Our goal is to evaluate forecasting accuracy at the individual participant level and compare it with the accuracy obtained from forecasting on aggregated consumption/production data within a specific grid segment or using data from the secondary substation to which the participants are connected. |
| 8. | Andreas Kontogiannis Vasilis Pollatos, Gabriele Farina Panayotis Mertikopoulos Ioannis Panageas : Efficient kernelized learning in polyhedral games beyond full-information: From Colonel Blotto to congestion games. The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025 poster), 2025. (Type: Conference | Abstract | Links | BibTeX) @conference{Kontogiannis2025b, title = {Efficient kernelized learning in polyhedral games beyond full-information: From Colonel Blotto to congestion games}, author = {Andreas Kontogiannis, Vasilis Pollatos, Gabriele Farina, Panayotis Mertikopoulos, Ioannis Panageas}, url = {https://openreview.net/attachment?id=FUBaZDMOFj&name=pdf https://arxiv.org/pdf/2509.20919}, doi = {https://doi.org/10.48550/arXiv.2509.20919}, year = {2025}, date = {2025-09-25}, booktitle = {The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025 poster)}, journal = {The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025 poster)}, abstract = {We examine the problem of efficiently learning coarse correlated equilibria (CCE) in polyhedral games, that is, normal-form games with an exponentially large number of actions per player and an underlying combinatorial structure—such as the classic Colonel Blotto game or congestion games. Achieving computational efficiency in this setting requires learning algorithms whose regret and per-iteration complexity scale at most polylogarithmically with the size of the players’ action sets. This challenge has recently been addressed in the full-information setting, primarily through the use of kernelization; however, in the more realistic partial information setting, the situation is much more challenging, and existing approaches result in suboptimal and impractical runtime complexity to learn CCE. We address this gap via a novel kernelization-based framework for payoff-based learning in polyhedral games, which we then apply to certain key classes of polyhedral games—namely Colonel Blotto, graphic matroid and network congestion games. In so doing, we obtain a range of computationally efficient payoff-based learning algorithms which significantly improve upon prior work in terms of the runtime for learning CCE.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } We examine the problem of efficiently learning coarse correlated equilibria (CCE) in polyhedral games, that is, normal-form games with an exponentially large number of actions per player and an underlying combinatorial structure—such as the classic Colonel Blotto game or congestion games. Achieving computational efficiency in this setting requires learning algorithms whose regret and per-iteration complexity scale at most polylogarithmically with the size of the players’ action sets. This challenge has recently been addressed in the full-information setting, primarily through the use of kernelization; however, in the more realistic partial information setting, the situation is much more challenging, and existing approaches result in suboptimal and impractical runtime complexity to learn CCE. We address this gap via a novel kernelization-based framework for payoff-based learning in polyhedral games, which we then apply to certain key classes of polyhedral games—namely Colonel Blotto, graphic matroid and network congestion games. In so doing, we obtain a range of computationally efficient payoff-based learning algorithms which significantly improve upon prior work in terms of the runtime for learning CCE. |
| 9. | Theocharis Kravaris, George Vouros A: Transferable aircraft trajectory prediction with generative deep imitation learning. In: International Journal of Data Science and Analytics, 20 (3), pp. 1977-1999, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Kravaris2025, title = {Transferable aircraft trajectory prediction with generative deep imitation learning}, author = {Theocharis Kravaris, George A Vouros}, url = {https://link.springer.com/article/10.1007/s41060-024-00574-1}, doi = {https://doi.org/10.1007/s41060-024-00574-1}, year = {2025}, date = {2025-09-01}, journal = {International Journal of Data Science and Analytics}, volume = {20}, number = {3}, pages = {1977-1999}, abstract = {Trajectory-oriented transformations to air traffic management operations require high fidelity aircraft trajectory prediction capabilities. Data-driven trajectory prediction approaches provide promising results, albeit with important limitations that hinder seriously the efficient and effective deployment of trajectory prediction methods: They need abundant training effort with a large amount of training samples and require training distinct models for different origin–destination (OD) airport pairs. In this paper, we address the problem of building transferable trajectory prediction models, casting the prediction problem as a transferable imitation task, introducing a novel formulation which (a) provides the capability to utilize trained models, in new OD pairs, offering a warm starting for computationally efficient training, and (b) improves the efficacy of data-driven trajectory prediction. The proposed approach provides very accurate results for large look-ahead time predictions, even if transferable models have been trained with few samples.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Trajectory-oriented transformations to air traffic management operations require high fidelity aircraft trajectory prediction capabilities. Data-driven trajectory prediction approaches provide promising results, albeit with important limitations that hinder seriously the efficient and effective deployment of trajectory prediction methods: They need abundant training effort with a large amount of training samples and require training distinct models for different origin–destination (OD) airport pairs. In this paper, we address the problem of building transferable trajectory prediction models, casting the prediction problem as a transferable imitation task, introducing a novel formulation which (a) provides the capability to utilize trained models, in new OD pairs, offering a warm starting for computationally efficient training, and (b) improves the efficacy of data-driven trajectory prediction. The proposed approach provides very accurate results for large look-ahead time predictions, even if transferable models have been trained with few samples. |
| 10. | George Papadopoulos, George Vouros A: Learning safe, constrained policies via imitation learning: Connection to Probabilistic Inference and a Naive Algorithm. In: arXiv, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Papadopoulos2025b, title = {Learning safe, constrained policies via imitation learning: Connection to Probabilistic Inference and a Naive Algorithm}, author = {George Papadopoulos, George A Vouros}, url = {https://arxiv.org/pdf/2507.06780}, doi = {https://doi.org/10.48550/arXiv.2507.06780}, year = {2025}, date = {2025-07-09}, journal = {arXiv}, abstract = {This article introduces an imitation learning method for learning maximum entropy policies that comply with constraints demonstrated by expert trajectories executing a task. The formulation of the method takes advantage of results connecting performance to bounds for the KL-divergence between demonstrated and learned policies, and its objective is rigorously justified through a connection to a probabilistic inference framework for reinforcement learning, incorporating the reinforcement learning objective and the objective to abide by constraints in an entropy maximization setting. The proposed algorithm optimizes the learning objective with dual gradient descent, supporting effective and stable training. Experiments show that the proposed method can learn effective policy models for constraints-abiding behaviour, in settings with multiple constraints of different types, accommodating different modalities of demonstrated behaviour, and with abilities to generalize.}, keywords = {}, pubstate = {published}, tppubtype = {article} } This article introduces an imitation learning method for learning maximum entropy policies that comply with constraints demonstrated by expert trajectories executing a task. The formulation of the method takes advantage of results connecting performance to bounds for the KL-divergence between demonstrated and learned policies, and its objective is rigorously justified through a connection to a probabilistic inference framework for reinforcement learning, incorporating the reinforcement learning objective and the objective to abide by constraints in an entropy maximization setting. The proposed algorithm optimizes the learning objective with dual gradient descent, supporting effective and stable training. Experiments show that the proposed method can learn effective policy models for constraints-abiding behaviour, in settings with multiple constraints of different types, accommodating different modalities of demonstrated behaviour, and with abilities to generalize. |
| 11. | Piyabhum Chaysri Theodoros Tranos, George Papadopoulos George Vouros Konstantinos Blekas A: Efficient Autonomous Marine Vessel Navigation with Safe Deep Reinforcement Learning. 2025 Symposium on Maritime Informatics and Robotics (MARIS), 2025. (Type: Conference | Abstract | Links | BibTeX) @conference{Chaysri2025, title = { Efficient Autonomous Marine Vessel Navigation with Safe Deep Reinforcement Learning}, author = {Piyabhum Chaysri, Theodoros Tranos, George Papadopoulos, George A Vouros, Konstantinos Blekas}, doi = {https://doi.org/10.1109/MARIS64137.2025.11139786}, year = {2025}, date = {2025-06-26}, booktitle = {2025 Symposium on Maritime Informatics and Robotics (MARIS)}, abstract = {The rise of automation and self-driving systems brings a strong focus on safety-centric decision-making, especially in complex environments with large degree of uncertainty where unpredictable interactions occur at high frequency. In this study we address the challenge of safe and efficient maritime navigation by proposing a safe Deep Reinforcement Learning scheme for training Unmanned Surface Vehicle (USV) agents. Our approach leverages the Lagrangian relaxation framework to effectively handle safety constraints, ensuring that the learned navigation policies balance goal achievement with obstacle avoidance. We address each type of static and moving obstacle separately with the aim of achieving more effective management of their impact on safe navigation. This enables the design of a more advanced constraint-aware optimization framework, enhancing USV’s ability to navigate complex maritime environment, adapt to changing traffic conditions and maintain a minimal risk of collision. Experiments were conducted in a simulated environment tailored to match realistic weather and traffic density conditions. The simulation results highlight the potential of the proposed method in developing advanced USV navigation policies that achieve high accuracy and enhanced safety.}, keywords = {}, pubstate = {published}, tppubtype = {conference} } The rise of automation and self-driving systems brings a strong focus on safety-centric decision-making, especially in complex environments with large degree of uncertainty where unpredictable interactions occur at high frequency. In this study we address the challenge of safe and efficient maritime navigation by proposing a safe Deep Reinforcement Learning scheme for training Unmanned Surface Vehicle (USV) agents. Our approach leverages the Lagrangian relaxation framework to effectively handle safety constraints, ensuring that the learned navigation policies balance goal achievement with obstacle avoidance. We address each type of static and moving obstacle separately with the aim of achieving more effective management of their impact on safe navigation. This enables the design of a more advanced constraint-aware optimization framework, enhancing USV’s ability to navigate complex maritime environment, adapt to changing traffic conditions and maintain a minimal risk of collision. Experiments were conducted in a simulated environment tailored to match realistic weather and traffic density conditions. The simulation results highlight the potential of the proposed method in developing advanced USV navigation policies that achieve high accuracy and enhanced safety. |
| 12. | George Giannakopoulos Andreas Sideras, Konstantinos Stamatakis Nikolaos Melanitis : NAVMAT: An AI-supported naval failures knowledge management system. In: Expert Systems with Applications, 277 , pp. 127117, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Giannakopoulos2025, title = {NAVMAT: An AI-supported naval failures knowledge management system}, author = {George Giannakopoulos, Andreas Sideras, Konstantinos Stamatakis, Nikolaos Melanitis}, doi = {https://doi.org/10.1016/j.eswa.2025.127117}, year = {2025}, date = {2025-06-05}, journal = {Expert Systems with Applications}, volume = {277}, pages = {127117}, abstract = {We present “NAVMAT”, an intelligent, multilingual knowledge management platform designed to record and categorize material failure incidents reported in naval operations. This paper provides an overview of the platform, identifying its key software components and highlighting the information retrieval approach used to support user workflows. The platform primarily facilitates real-time, multilingual search and intelligent indexing, streamlining the incident management process while offering valuable insights from past incidents and knowledge resources. To achieve this, it employs a customized natural language processing pipeline integrated with a carefully engineered ontology. The ontology, regularly updated by domain experts, enriches the retrieval mechanism by instilling domain specific knowledge. This approach aims to reduce the significant variability in specialized terminology by promoting convergence towards a unified vocabulary.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We present “NAVMAT”, an intelligent, multilingual knowledge management platform designed to record and categorize material failure incidents reported in naval operations. This paper provides an overview of the platform, identifying its key software components and highlighting the information retrieval approach used to support user workflows. The platform primarily facilitates real-time, multilingual search and intelligent indexing, streamlining the incident management process while offering valuable insights from past incidents and knowledge resources. To achieve this, it employs a customized natural language processing pipeline integrated with a carefully engineered ontology. The ontology, regularly updated by domain experts, enriches the retrieval mechanism by instilling domain specific knowledge. This approach aims to reduce the significant variability in specialized terminology by promoting convergence towards a unified vocabulary. |
| 13. | Andreas Kontogiannis Konstantinos Papathanasiou, Yi Shen Giorgos Stamou Michael Zavlanos George Vouros M: Enhancing cooperative multi-agent reinforcement learning with state modelling and adversarial exploration. In: arXiv, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Kontogiannis2025, title = {Enhancing cooperative multi-agent reinforcement learning with state modelling and adversarial exploration}, author = {Andreas Kontogiannis, Konstantinos Papathanasiou, Yi Shen, Giorgos Stamou, Michael M Zavlanos, George Vouros}, url = {https://arxiv.org/pdf/2505.05262}, doi = {https://doi.org/10.48550/arXiv.2505.05262}, year = {2025}, date = {2025-05-08}, journal = {arXiv}, abstract = {Learning to cooperate in distributed partially observable environments with no communication abilities poses significant challenges for multi-agent deep reinforcement learning (MARL). This paper addresses key concerns in this domain, focusing on inferring state representations from individual agent observations and leveraging these representations to enhance agents’ exploration and collaborative task execution policies. To this end, we propose a novel state modelling framework for cooperative MARL, where agents infer meaningful belief representations of the non-observable state, with respect to optimizing their own policies, while filtering redundant and less informative joint state information. Building upon this framework, we propose the MARL SMPE algorithm. In SMPE, agents enhance their own policy’s discriminative abilities under partial observability, explicitly by incorporating their beliefs into the policy network, and implicitly by adopting an adversarial type of exploration policies which encourages agents to discover novel, high-value states while improving the discriminative abilities of others. Experimentally, we show that SMPE outperforms state-of-the-art MARL algorithms in complex fully cooperative tasks from the MPE, LBF, and RWARE benchmarks.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Learning to cooperate in distributed partially observable environments with no communication abilities poses significant challenges for multi-agent deep reinforcement learning (MARL). This paper addresses key concerns in this domain, focusing on inferring state representations from individual agent observations and leveraging these representations to enhance agents’ exploration and collaborative task execution policies. To this end, we propose a novel state modelling framework for cooperative MARL, where agents infer meaningful belief representations of the non-observable state, with respect to optimizing their own policies, while filtering redundant and less informative joint state information. Building upon this framework, we propose the MARL SMPE algorithm. In SMPE, agents enhance their own policy’s discriminative abilities under partial observability, explicitly by incorporating their beliefs into the policy network, and implicitly by adopting an adversarial type of exploration policies which encourages agents to discover novel, high-value states while improving the discriminative abilities of others. Experimentally, we show that SMPE outperforms state-of-the-art MARL algorithms in complex fully cooperative tasks from the MPE, LBF, and RWARE benchmarks. |
| 14. | Despoina P Kiouri Georgios C Batsis, Thomas Mavromoustakos Alessandro Giuliani Christos Chasapis T: Structure-Based Modeling of the Gut Bacteria–Host Interactome Through Statistical Analysis of Domain–Domain Associations Using Machine Learning. In: BioTech, 14 (1), pp. 13, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Kiouri2025c, title = {Structure-Based Modeling of the Gut Bacteria–Host Interactome Through Statistical Analysis of Domain–Domain Associations Using Machine Learning}, author = {Despoina P Kiouri, Georgios C Batsis, Thomas Mavromoustakos, Alessandro Giuliani, Christos T Chasapis}, url = {https://www.mdpi.com/2673-6284/14/1/13/pdf?version=1740479495}, doi = {https://doi.org/10.3390/biotech14010013}, year = {2025}, date = {2025-02-25}, journal = {BioTech}, volume = {14}, number = {1}, pages = {13}, abstract = {The gut microbiome, a complex ecosystem of microorganisms, plays a pivotal role in human health and disease. The gut microbiome’s influence extends beyond the digestive system to various organs, and its imbalance is linked to a wide range of diseases, including cancer and neurodevelopmental, inflammatory, metabolic, cardiovascular, autoimmune, and psychiatric diseases. Despite its significance, the interactions between gut bacteria and human proteins remain understudied, with less than 20,000 experimentally validated protein interactions between the host and any bacteria species. This study addresses this knowledge gap by predicting a protein–protein interaction network between gut bacterial and human proteins. Using statistical associations between Pfam domains, a comprehensive dataset of over one million experimentally validated pan-bacterial–human protein interactions, as well as inter- and intra-species protein interactions from various organisms, were used for the development of a machine learning-based prediction method to uncover key regulatory molecules in this dynamic system. This study’s findings contribute to the understanding of the intricate gut microbiome–host relationship and pave the way for future experimental validation and therapeutic strategies targeting the gut microbiome interplay.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The gut microbiome, a complex ecosystem of microorganisms, plays a pivotal role in human health and disease. The gut microbiome’s influence extends beyond the digestive system to various organs, and its imbalance is linked to a wide range of diseases, including cancer and neurodevelopmental, inflammatory, metabolic, cardiovascular, autoimmune, and psychiatric diseases. Despite its significance, the interactions between gut bacteria and human proteins remain understudied, with less than 20,000 experimentally validated protein interactions between the host and any bacteria species. This study addresses this knowledge gap by predicting a protein–protein interaction network between gut bacterial and human proteins. Using statistical associations between Pfam domains, a comprehensive dataset of over one million experimentally validated pan-bacterial–human protein interactions, as well as inter- and intra-species protein interactions from various organisms, were used for the development of a machine learning-based prediction method to uncover key regulatory molecules in this dynamic system. This study’s findings contribute to the understanding of the intricate gut microbiome–host relationship and pave the way for future experimental validation and therapeutic strategies targeting the gut microbiome interplay. |
| 15. | Natalia Koliou, George Vouros : Ranking Joint Policies in Dynamic Games using Evolutionary Dynamics. In: arXiv, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Koliou2025, title = {Ranking Joint Policies in Dynamic Games using Evolutionary Dynamics}, author = {Natalia Koliou, George Vouros}, url = {https://arxiv.org/pdf/2502.14724}, doi = {https://doi.org/10.48550/arXiv.2502.14724}, year = {2025}, date = {2025-02-20}, journal = {arXiv}, abstract = {Game-theoretic solution concepts, such as the Nash equilibrium, have been key to finding stable joint actions in multi-player games. However, it has been shown that the dynamics of agents’ interactions, even in simple two-player games with few strategies, are incapable of reaching Nash equilibria, exhibiting complex and unpredictable behavior. Instead, evolutionary approaches can describe the long-term persistence of strategies and filter out transient ones, accounting for the long-term dynamics of agents’ interactions. Our goal is to identify agents’ joint strategies that result in stable behavior, being resistant to changes, while also accounting for agents’ payoffs, in dynamic games. Towards this goal, and building on previous results, this paper proposes transforming dynamic games into their empirical forms by considering agents’ strategies instead of agents’ actions, and applying the evolutionary methodology -Rank to evaluate and rank strategy profiles according to their long-term dynamics. This methodology not only allows us to identify joint strategies that are strong through agents’ long-term interactions, but also provides a descriptive, transparent framework regarding the high ranking of these strategies. Experiments report on agents that aim to collaboratively solve a stochastic version of the graph coloring problem. We consider different styles of play as strategies to define the empirical game, and train policies realizing these strategies, using the DQN algorithm. Then we run simulations to generate the payoff matrix required by -Rank to rank joint strategies.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Game-theoretic solution concepts, such as the Nash equilibrium, have been key to finding stable joint actions in multi-player games. However, it has been shown that the dynamics of agents’ interactions, even in simple two-player games with few strategies, are incapable of reaching Nash equilibria, exhibiting complex and unpredictable behavior. Instead, evolutionary approaches can describe the long-term persistence of strategies and filter out transient ones, accounting for the long-term dynamics of agents’ interactions. Our goal is to identify agents’ joint strategies that result in stable behavior, being resistant to changes, while also accounting for agents’ payoffs, in dynamic games. Towards this goal, and building on previous results, this paper proposes transforming dynamic games into their empirical forms by considering agents’ strategies instead of agents’ actions, and applying the evolutionary methodology -Rank to evaluate and rank strategy profiles according to their long-term dynamics. This methodology not only allows us to identify joint strategies that are strong through agents’ long-term interactions, but also provides a descriptive, transparent framework regarding the high ranking of these strategies. Experiments report on agents that aim to collaboratively solve a stochastic version of the graph coloring problem. We consider different styles of play as strategies to define the empirical game, and train policies realizing these strategies, using the DQN algorithm. Then we run simulations to generate the payoff matrix required by -Rank to rank joint strategies. |
| 16. | Christos Spatharis Konstantinos Blekas, George Santipantakis George Vouros : Modular and Multimodal Generative Adversarial Imitation Learning for Modeling Flight Trajectories. In: Journal of Air Transportation, 33 (3), pp. 188-204, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Spatharis2025, title = {Modular and Multimodal Generative Adversarial Imitation Learning for Modeling Flight Trajectories}, author = {Christos Spatharis, Konstantinos Blekas, George Santipantakis, George Vouros}, doi = {https://doi.org/10.2514/1.D0396}, year = {2025}, date = {2025-02-17}, journal = {Journal of Air Transportation}, volume = {33}, number = {3}, pages = {188-204}, abstract = {We aim to imitate the execution of modular tasks by exploiting unsegmented trajectories that demonstrate the execution of these tasks. This is challenging since the execution of tasks follows different modes (i.e., patterns of behavior), which may exist in various mixtures within subtasks, and the identification of trajectories’ modules (i.e., subtrajectories executing subtasks) may not be easy. This paper addresses the modularity of trajectories in conjunction with multimodality toward imitating the execution of aircraft trajectories. It proposes an imitation learning framework for the aircraft trajectory prediction problem, which segments demonstrated aircraft trajectories into subtrajectories corresponding to flight phases. This facilitates disentangling modes and learning a mixture of policies per flight phase. While trajectories are segmented using domain-specific rules, a mixture of policies per flight phase is learned by a generative multimodal imitation learning method. This modular approach enables accurate prediction of both modes and subtrajectories, which finally results in predicting the evolution of the aircraft state across the whole trajectory in a compositional way. Experiments using a real-world dataset of long flights show the potential of the proposed framework to disentangle multimodal trajectories in real-world settings and predict trajectories with high accuracy, in comparison to methods that do not exploit subtrajectories.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We aim to imitate the execution of modular tasks by exploiting unsegmented trajectories that demonstrate the execution of these tasks. This is challenging since the execution of tasks follows different modes (i.e., patterns of behavior), which may exist in various mixtures within subtasks, and the identification of trajectories’ modules (i.e., subtrajectories executing subtasks) may not be easy. This paper addresses the modularity of trajectories in conjunction with multimodality toward imitating the execution of aircraft trajectories. It proposes an imitation learning framework for the aircraft trajectory prediction problem, which segments demonstrated aircraft trajectories into subtrajectories corresponding to flight phases. This facilitates disentangling modes and learning a mixture of policies per flight phase. While trajectories are segmented using domain-specific rules, a mixture of policies per flight phase is learned by a generative multimodal imitation learning method. This modular approach enables accurate prediction of both modes and subtrajectories, which finally results in predicting the evolution of the aircraft state across the whole trajectory in a compositional way. Experiments using a real-world dataset of long flights show the potential of the proposed framework to disentangle multimodal trajectories in real-world settings and predict trajectories with high accuracy, in comparison to methods that do not exploit subtrajectories. |
| 17. | Despoina P Kiouri Georgios C Batsis, Christos Chasapis T: Structure-Based Deep Learning Framework for Modeling Human–Gut Bacterial Protein Interactions. In: Proteomes, 13 (1), pp. 10, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Kiouri2025b, title = {Structure-Based Deep Learning Framework for Modeling Human–Gut Bacterial Protein Interactions}, author = {Despoina P Kiouri, Georgios C Batsis, Christos T Chasapis}, url = {https://www.mdpi.com/2227-7382/13/1/10/pdf?version=1739790880}, doi = {https://doi.org/10.3390/proteomes13010010}, year = {2025}, date = {2025-02-17}, journal = {Proteomes}, volume = {13}, number = {1}, pages = {10}, abstract = {The interaction network between the human host proteins and the proteins of the gut bacteria is essential for the establishment of human health, and its dysregulation directly contributes to disease development. Despite its great importance, experimental data on protein–protein interactions (PPIs) between these species are sparse due to experimental limitations. Methods: This study presents a deep learning-based framework for predicting PPIs between human and gut bacterial proteins using structural data. The framework leverages graph-based protein representations and variational autoencoders (VAEs) to extract structural embeddings from protein graphs, which are then fused through a Bi-directional Cross-Attention module to predict interactions. The model addresses common challenges in PPI datasets, such as class imbalance, using focal loss to emphasize harder-to-classify samples. Results: The results demonstrated that this framework exhibits robust performance, with high precision and recall across validation and test datasets, underscoring its generalizability. By incorporating proteoforms in the analysis, the model accounts for the structural complexity within proteomes, making predictions biologically relevant. Conclusions: These findings offer a scalable tool for investigating the interactions between the host and the gut microbiota, potentially yielding new treatment targets and diagnostics for disorders linked to the microbiome.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The interaction network between the human host proteins and the proteins of the gut bacteria is essential for the establishment of human health, and its dysregulation directly contributes to disease development. Despite its great importance, experimental data on protein–protein interactions (PPIs) between these species are sparse due to experimental limitations. Methods: This study presents a deep learning-based framework for predicting PPIs between human and gut bacterial proteins using structural data. The framework leverages graph-based protein representations and variational autoencoders (VAEs) to extract structural embeddings from protein graphs, which are then fused through a Bi-directional Cross-Attention module to predict interactions. The model addresses common challenges in PPI datasets, such as class imbalance, using focal loss to emphasize harder-to-classify samples. Results: The results demonstrated that this framework exhibits robust performance, with high precision and recall across validation and test datasets, underscoring its generalizability. By incorporating proteoforms in the analysis, the model accounts for the structural complexity within proteomes, making predictions biologically relevant. Conclusions: These findings offer a scalable tool for investigating the interactions between the host and the gut microbiota, potentially yielding new treatment targets and diagnostics for disorders linked to the microbiome. |
| 18. | George Papadopoulos Andreas Kontogiannis, Foteini Papadopoulou Chaido Poulianou Ioannis Koumentis George Vouros : An extended benchmarking of multi-agent reinforcement learning algorithms in complex fully cooperative tasks. In: arXiv, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Papadopoulos2025, title = {An extended benchmarking of multi-agent reinforcement learning algorithms in complex fully cooperative tasks}, author = {George Papadopoulos, Andreas Kontogiannis, Foteini Papadopoulou, Chaido Poulianou, Ioannis Koumentis, George Vouros}, url = {https://arxiv.org/pdf/2502.04773}, doi = {https://doi.org/10.48550/arXiv.2502.04773}, year = {2025}, date = {2025-02-07}, journal = {arXiv}, abstract = {Multi-Agent Reinforcement Learning (MARL) has recently emerged as a significant area of research. However, MARL evaluation often lacks systematic diversity, hindering a comprehensive understanding of algorithms’ capabilities. In particular, cooperative MARL algorithms are predominantly evaluated on benchmarks such as SMAC and GRF, which primarily feature team game scenarios without assessing adequately various aspects of agents’ capabilities required in fully cooperative real-world tasks such as multi-robot cooperation and warehouse, resource management, search and rescue, and human-AI cooperation. Moreover, MARL algorithms are mainly evaluated on low dimensional state spaces, and thus their performance on high-dimensional (e.g., image) observations is not well-studied. To fill this gap, this paper highlights the crucial need for expanding systematic evaluation across a wider array of existing benchmarks. To this end, we conduct extensive evaluation and comparisons of well-known MARL algorithms on complex fully cooperative benchmarks, including tasks with images as agents’ observations. Interestingly, our analysis shows that many algorithms, hailed as state-of-the-art on SMAC and GRF, may underperform standard MARL baselines on fully cooperative benchmarks. Finally, towards more systematic and better evaluation of cooperative MARL algorithms, we have open-sourced PyMARLzoo+, an extension of the widely used (E)PyMARL libraries, which addresses an open challenge from [49], facilitating seamless integration and support with all benchmarks of PettingZoo, as well as Overcooked, PressurePlate, Capture Target and Box Pushing.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Multi-Agent Reinforcement Learning (MARL) has recently emerged as a significant area of research. However, MARL evaluation often lacks systematic diversity, hindering a comprehensive understanding of algorithms’ capabilities. In particular, cooperative MARL algorithms are predominantly evaluated on benchmarks such as SMAC and GRF, which primarily feature team game scenarios without assessing adequately various aspects of agents’ capabilities required in fully cooperative real-world tasks such as multi-robot cooperation and warehouse, resource management, search and rescue, and human-AI cooperation. Moreover, MARL algorithms are mainly evaluated on low dimensional state spaces, and thus their performance on high-dimensional (e.g., image) observations is not well-studied. To fill this gap, this paper highlights the crucial need for expanding systematic evaluation across a wider array of existing benchmarks. To this end, we conduct extensive evaluation and comparisons of well-known MARL algorithms on complex fully cooperative benchmarks, including tasks with images as agents’ observations. Interestingly, our analysis shows that many algorithms, hailed as state-of-the-art on SMAC and GRF, may underperform standard MARL baselines on fully cooperative benchmarks. Finally, towards more systematic and better evaluation of cooperative MARL algorithms, we have open-sourced PyMARLzoo+, an extension of the widely used (E)PyMARL libraries, which addresses an open challenge from [49], facilitating seamless integration and support with all benchmarks of PettingZoo, as well as Overcooked, PressurePlate, Capture Target and Box Pushing. |
| 19. | Fotis Assimakopoulos Costas Vassilakis, Dionisis Margaris Konstantinos Kotis Dimitris Spiliotopoulos : AI and related technologies in the fields of smart agriculture: A review. In: Information, 16 (2), pp. 100, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Assimakopoulos2025, title = {AI and related technologies in the fields of smart agriculture: A review}, author = {Fotis Assimakopoulos, Costas Vassilakis, Dionisis Margaris, Konstantinos Kotis, Dimitris Spiliotopoulos}, url = {https://www.mdpi.com/2078-2489/16/2/100/pdf?version=1738918054}, doi = {https://doi.org/10.3390/info16020100}, year = {2025}, date = {2025-02-02}, journal = {Information}, volume = {16}, number = {2}, pages = {100}, abstract = {The integration of cutting-edge technologies—such as the Internet of Things (IoT), artificial intelligence (AI), machine learning (ML), and various emerging technologies—is revolutionizing agricultural practices, enhancing productivity, sustainability, and efficiency. The objective of this study is to review the literature regarding the development and evolution of AI as well as other emerging technologies in the various fields of Agriculture as they are developed and transformed by integrating the above technologies. The areas examined in this study are open field smart farming, vertical and indoor farming, zero waste agriculture, precision livestock farming, smart greenhouses, and regenerative agriculture. This paper links current research, technological innovations, and case studies to present a comprehensive review of these emerging technologies being developed in the context of smart agriculture, for the benefit of farmers and consumers in general. By exploring practical applications and future perspectives, this work aims to provide valuable insights to address global food security challenges, minimize environmental impacts, and support sustainable development goals through the application of new technologies.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The integration of cutting-edge technologies—such as the Internet of Things (IoT), artificial intelligence (AI), machine learning (ML), and various emerging technologies—is revolutionizing agricultural practices, enhancing productivity, sustainability, and efficiency. The objective of this study is to review the literature regarding the development and evolution of AI as well as other emerging technologies in the various fields of Agriculture as they are developed and transformed by integrating the above technologies. The areas examined in this study are open field smart farming, vertical and indoor farming, zero waste agriculture, precision livestock farming, smart greenhouses, and regenerative agriculture. This paper links current research, technological innovations, and case studies to present a comprehensive review of these emerging technologies being developed in the context of smart agriculture, for the benefit of farmers and consumers in general. By exploring practical applications and future perspectives, this work aims to provide valuable insights to address global food security challenges, minimize environmental impacts, and support sustainable development goals through the application of new technologies. |
| 20. | Dimitrios Doumanas Georgios Bouchouras, Andreas Soularidis Konstantinos Kotis George Vouros : From human-to LLM-centered collaborative ontology engineering. In: Applied Ontology, 19 (4), pp. 334-367, 2025. (Type: Journal Article | Abstract | Links | BibTeX) @article{Doumanas2025, title = {From human-to LLM-centered collaborative ontology engineering}, author = {Dimitrios Doumanas, Georgios Bouchouras, Andreas Soularidis, Konstantinos Kotis, George Vouros}, doi = {https://doi.org/10.1177/15705838241305067}, year = {2025}, date = {2025-01-31}, journal = {Applied Ontology}, volume = {19}, number = {4}, pages = {334-367}, abstract = {In the continuously evolving landscape of knowledge engineering, the symbiosis and teaming of humans and machines emerge as a pivotal new domain. This article explores the multifaceted realms of human and machine collaborative ontology engineering (OE). The goal of the presented work is to explore the potential of Large Language Models (LLMs) to speed up and automate the processes of collaborative OE, experimenting with different levels of LLM involvement. The proposed approach is based on a human-centered approach, that is, the HCOME approach to collaborative OE, and follows a process of exploring the declining involvement of humans and the parallel increase of LLM involvement, concluding at a level of automation where the OE is exclusively performed by LLMs. This experimentation is organized based on a series of human/LLM collaboration levels (a spectrum of OE), each one aligned to a specific OE methodology, that is, Level-0 HCOME (Human), Level-1 X-HCOME (Human and LLMs), Level-2 SimX-HCOME (LLMs and Human), and Level-3 Sim-HCOME (LLMs). The evaluation of these methodologies (one per level) is performed by measuring the similarity of the generated ontologies against “reference” ontologies (precision, recall, and F1-score of reference-to-LLM-generated ontological mappings). The results presented in this paper demonstrate that while LLMs significantly expedite the OE process, the accuracy and completeness of the resulting ontologies are notably enhanced by maintaining a high level of human involvement. This study is expected to contribute to a deeper understanding of evolving dynamics in LLM-based/enhanced OE, paving the way for future advancements toward more effective collaborative OE frameworks.}, keywords = {}, pubstate = {published}, tppubtype = {article} } In the continuously evolving landscape of knowledge engineering, the symbiosis and teaming of humans and machines emerge as a pivotal new domain. This article explores the multifaceted realms of human and machine collaborative ontology engineering (OE). The goal of the presented work is to explore the potential of Large Language Models (LLMs) to speed up and automate the processes of collaborative OE, experimenting with different levels of LLM involvement. The proposed approach is based on a human-centered approach, that is, the HCOME approach to collaborative OE, and follows a process of exploring the declining involvement of humans and the parallel increase of LLM involvement, concluding at a level of automation where the OE is exclusively performed by LLMs. This experimentation is organized based on a series of human/LLM collaboration levels (a spectrum of OE), each one aligned to a specific OE methodology, that is, Level-0 HCOME (Human), Level-1 X-HCOME (Human and LLMs), Level-2 SimX-HCOME (LLMs and Human), and Level-3 Sim-HCOME (LLMs). The evaluation of these methodologies (one per level) is performed by measuring the similarity of the generated ontologies against “reference” ontologies (precision, recall, and F1-score of reference-to-LLM-generated ontological mappings). The results presented in this paper demonstrate that while LLMs significantly expedite the OE process, the accuracy and completeness of the resulting ontologies are notably enhanced by maintaining a high level of human involvement. This study is expected to contribute to a deeper understanding of evolving dynamics in LLM-based/enhanced OE, paving the way for future advancements toward more effective collaborative OE frameworks. |