Benchmarking Perturbation-Based Saliency Maps for Explaining Atari Agents

One of the most prominent methods for explaining the behavior of Deep Reinforcement Learning (DRL) agents is the generation of saliency maps that show how much each pixel attributed to the agents' decision. However, there is no work that computationally evaluates and compares the fidelity of differe...

Πλήρης περιγραφή

Λεπτομέρειες βιβλιογραφικής εγγραφής

Κύριοι συγγραφείς:	Tobias Huber, Benedikt Limmer, Elisabeth André
Μορφή:	Artigo
Γλώσσα:	Inglês
Έκδοση:	Frontiers Media S.A. 2022-07-01
Σειρά:	Frontiers in Artificial Intelligence
Θέματα:	explainable reinforcement learning explainable artificial intelligence (XAI) interpretable machine learning deep reinforcement learning feature attribution saliency maps
Διαθέσιμο Online:	https://www.frontiersin.org/articles/10.3389/frai.2022.903875/full
Ετικέτες:	Προσθήκη ετικέτας Δεν υπάρχουν, Καταχωρήστε ετικέτα πρώτοι!