George A. Vouros

George A. Vouros

Interpretability, explainability and transparency are key issues to introducing Artificial Intelligence methods in many critical domains: This is important due to ethical concerns and trust issues strongly connected to reliability, robustness, auditability and fairness, and has important consequences towards keeping the human in the loop in high levels of automation, especially in critical cases for decision making, where both (human and the machine) play important roles. While the research community has given much attention to explainability of closed (or black) prediction boxes, there are tremendous needs for explainability of closed-box methods that support agents to act autonomously in the real world. Reinforcement learning methods, and especially their deep versions, are such closed-box methods. In this article we aim to provide a review of state of the art methods for explainable deep reinforcement learning methods, taking also into account the needs of human operators - i.e., of those that take the actual and critical decisions in solving real-world problems. We provide a formal specification of the deep reinforcement learning explainability problems, and we identify the necessary components of a general explainable reinforcement learning framework. Based on these, we provide a comprehensive review of state of the art methods, categorizing them in classes according to the paradigm they follow, the interpretable models they use, and the surface representation of explanations provided. The article concludes identifying open questions and important challenges.

Alevizos Bastas, George A. Vouros

With the aim to enhance automation in conflict detection and resolution (CD&R) tasks in the Air Traffic Management domain, in this paper we propose deep learning techniques (DL) that can learn models of Air Traffic Controllers' (ATCO) reactions in resolving conflicts that can violate separation minimum constraints among aircraft trajectories: This implies learning when the ATCO will react towards resolving a conflict, and how he/she will react. Timely reactions, to which this paper aims, focus on when do reactions happen, aiming to predict the trajectory points, as the trajectory evolves, that the ATCO issues a conflict resolution action, while also predicting the type of resolution action (if any). Towards this goal, the paper formulates the ATCO reactions prediction problem for CD&R, and presents DL methods that can model ATCO timely reactions and evaluates these methods in real-world data sets, showing their efficacy in prediction with very high accuracy.

George Vouros, George Papadopoulos, Alevizos Bastas et al.

Dense and complex air traffic scenarios require higher levels of automation than those exhibited by tactical conflict detection and resolution (CD\&R) tools that air traffic controllers (ATCO) use today. However, the air traffic control (ATC) domain, being safety critical, requires AI systems to which operators are comfortable to relinquishing control, guaranteeing operational integrity and automation adoption. Two major factors towards this goal are quality of solutions, and transparency in decision making. This paper proposes using a graph convolutional reinforcement learning method operating in a multiagent setting where each agent (flight) performs a CD\&R task, jointly with other agents. We show that this method can provide high-quality solutions with respect to stakeholders interests (air traffic controllers and airspace users), addressing operational transparency issues.

Andreas Kontogiannis, George Vouros

Although deep reinforcement learning (DRL) methods have been successfully applied in challenging tasks, their application in real-world operational settings is challenged by methods' limited ability to provide explanations. Among the paradigms for explainability in DRL is the interpretable box design paradigm, where interpretable models substitute inner constituent models of the DRL method, thus making the DRL method "inherently" interpretable. In this paper we explore this paradigm and we propose XDQN, an explainable variation of DQN, which uses an interpretable policy model trained through mimicking. XDQN is challenged in a complex, real-world operational multi-agent problem, where agents are independent learners solving congestion problems. Specifically, XDQN is evaluated in three MARL scenarios, pertaining to the demand-capacity balancing problem of air traffic management. XDQN achieves performance similar to that of DQN, while its abilities to provide global models' interpretations and interpretations of local decisions are demonstrated.

Georgios Bouchouras, Dimitrios Doumanas, Andreas Soularidis et al.

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.

George Papadopoulos, Andreas Kontogiannis, Foteini Papadopoulou et al.

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 [TBG++21], facilitating seamless integration and support with all benchmarks of PettingZoo, as well as Overcooked, PressurePlate, Capture Target and Box Pushing.

George Papadopoulos, George A. Vouros

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.

Andreas Kontogiannis, Konstantinos Papathanasiou, Yi Shen et al.

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.

Natalia Koliou, George Vouros

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 $α$-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 $α$-Rank to rank joint strategies.

Alevizos Bastas, Theocharis Kravaris, George A. Vouros

The current Air Traffic Management (ATM) system worldwide has reached its limits in terms of predictability, efficiency and cost effectiveness. Different initiatives worldwide propose trajectory-oriented transformations that require high fidelity aircraft trajectory planning and prediction capabilities, supporting the trajectory life cycle at all stages efficiently. Recently proposed data-driven trajectory prediction approaches provide promising results. In this paper we approach the data-driven trajectory prediction problem as an imitation learning task, where we aim to imitate experts "shaping" the trajectory. Towards this goal we present a comprehensive framework comprising the Generative Adversarial Imitation Learning state of the art method, in a pipeline with trajectory clustering and classification methods. This approach, compared to other approaches, can provide accurate predictions for the whole trajectory (i.e. with a prediction horizon until reaching the destination) both at the pre-tactical (i.e. starting at the departure airport at a specific time instant) and at the tactical (i.e. from any state while flying) stages, compared to state of the art approaches.

Theocharis Kravaris, Christos Spatharis, Alevizos Bastas et al.

In this article, we report on the efficiency and effectiveness of multiagent reinforcement learning methods (MARL) for the computation of flight delays to resolve congestion problems in the Air Traffic Management (ATM) domain. Specifically, we aim to resolve cases where demand of airspace use exceeds capacity (demand-capacity problems), via imposing ground delays to flights at the pre-tactical stage of operations (i.e. few days to few hours before operation). Casting this into the multiagent domain, agents, representing flights, need to decide on own delays w.r.t. own preferences, having no information about others' payoffs, preferences and constraints, while they plan to execute their trajectories jointly with others, adhering to operational constraints. Specifically, we formalize the problem as a multiagent Markov Decision Process (MA-MDP) and we show that it can be considered as a Markov game in which interacting agents need to reach an equilibrium: What makes the problem more interesting is the dynamic setting in which agents operate, which is also due to the unforeseen, emergent effects of their decisions in the whole system. We propose collaborative multiagent reinforcement learning methods to resolve demand-capacity imbalances: Extensive experimental study on real-world cases, shows the potential of the proposed approaches in resolving problems, while advanced visualizations provide detailed views towards understanding the quality of solutions provided.

Konstantinos Kotis, Iraklis Athanasakis, George Vouros

In many cases, government data is still "locked" in several "data silos", even within the boundaries of a single (inter-)national public organization with disparate and distributed organizational units and departments spread across multiple sites. Opening data and enabling its unified querying from a single site in an efficient and effective way is a semantic application integration and open government data challenge. This paper describes how NARA is using Semantic Web technology to implement an application integration approach within the boundaries of its organization via opening and querying multiple governmental data sources from a single site. The generic approach proposed, namely S3-AI, provides support to answering unified, ontology-mediated, federated queries to data produced and exploited by disparate applications, while these are being located in different organizational sites. S3-AI preserves ownership, autonomy and independency of applications and data. The paper extensively demonstrates S3-AI, using the D2RQ and Fuseki technologies, for addressing the needs of a governmental "IT helpdesk support" case.

George A. Vouros, Georgios Santipantakis

Combining knowledge and beliefs of autonomous peers in distributed settings, is a ma- jor challenge. In this paper we consider peers that combine ontologies and reason jointly with their coupled knowledge. Ontologies are within the SHIQ fragment of Description Logics. Although there are several representation frameworks for modular Description Log- ics, each one makes crucial assumptions concerning the subjectivity of peers' knowledge, the relation between the domains over which ontologies are interpreted, the expressivity of the constructors used for combining knowledge, and the way peers share their knowledge. However in settings where autonomous peers can evolve and extend their knowledge and beliefs independently from others, these assumptions may not hold. In this article, we moti- vate the need for a representation framework that allows peers to combine their knowledge in various ways, maintaining the subjectivity of their own knowledge and beliefs, and that reason collaboratively, constructing a tableau that is distributed among them, jointly. The paper presents the proposed E-SHIQ representation framework, the implementation of the E-SHIQ distributed tableau reasoner, and discusses the efficiency of this reasoner.

Anastasios Skarlatidis, Georgios Paliouras, Alexander Artikis et al.

Symbolic event recognition systems have been successfully applied to a variety of application domains, extracting useful information in the form of events, allowing experts or other systems to monitor and respond when significant events are recognised. In a typical event recognition application, however, these systems often have to deal with a significant amount of uncertainty. In this paper, we address the issue of uncertainty in logic-based event recognition by extending the Event Calculus with probabilistic reasoning. Markov Logic Networks are a natural candidate for our logic-based formalism. However, the temporal semantics of the Event Calculus introduce a number of challenges for the proposed model. We show how and under what assumptions we can overcome these problems. Additionally, we study how probabilistic modelling changes the behaviour of the formalism, affecting its key property, the inertia of fluents. Furthermore, we demonstrate the advantages of the probabilistic Event Calculus through examples and experiments in the domain of activity recognition, using a publicly available dataset for video surveillance.