%0 Conference Proceedings
%T Predicate-based explanation of a Reinforcement Learning agent via action importance evaluation
%+ Université Toulouse III - Paul Sabatier (UT3)
%+ Argumentation, Décision, Raisonnement, Incertitude et Apprentissage (IRIT-ADRIA)
%A Saulières, Léo
%A Cooper, Martin
%A Dupin de Saint-Cyr, Florence
%< avec comité de lecture
%B 4th workshop on Advances in Interpretable Machine Learning and Artificial Intelligence (AIMLAI 2023) @ ECML/PKDD conference
%C Turin, Italy
%P à paraître
%8 2023-09-22
%D 2023
%K Explainable Artificial Intelligence
%K Reinforcement Learning
%K History Explanation
%Z Computer Science [cs]/Artificial Intelligence [cs.AI]Conference papers
%X For the purpose of understanding the impact of a Reinforcement Learning (RL) agent's decisions on the satisfaction of a given arbitrary predicate, we present a method based on the evaluation of the importance of actions. This highlights to the user the most important action(s) (relative to the predicate) in a history of the agent's interactions with the environment. Having shown that calculating the importance of an action for a predicate to hold is #W[1]-hard, we propose a timesaving approximation. To do so, we use the most likely transitions in the environment. Experiments confirm the relevance of this approach.
%G English
%2 https://hal.science/hal-04170188v1/document
%2 https://hal.science/hal-04170188v1/file/HXP_paper_for_AIMLAI_Workshop___Camera_Ready.pdf
%L hal-04170188
%U https://hal.science/hal-04170188
%~ UNIV-TLSE2
%~ UNIV-TLSE3
%~ CNRS
%~ UT1-CAPITOLE
%~ IRIT
%~ IRIT-ADRIA
%~ PNRIA
%~ ANR
%~ ANITI
%~ IRIT-IA
%~ IRIT-UT3
%~ TOULOUSE-INP
%~ UNIV-UT3
%~ UT3-INP
%~ UT3-TOULOUSEINP