Causal Explanation for Reinforcement Learning: Quantifying State and Temporal Importance
Artificial Intelligence
2023-07-04 v2 Machine Learning
Abstract
Explainability plays an increasingly important role in machine learning. Furthermore, humans view the world through a causal lens and thus prefer causal explanations over associational ones. Therefore, in this paper, we develop a causal explanation mechanism that quantifies the causal importance of states on actions and such importance over time. We also demonstrate the advantages of our mechanism over state-of-the-art associational methods in terms of RL policy explanation through a series of simulation studies, including crop irrigation, Blackjack, collision avoidance, and lunar lander.
Cite
@article{arxiv.2210.13507,
title = {Causal Explanation for Reinforcement Learning: Quantifying State and Temporal Importance},
author = {Xiaoxiao Wang and Fanyu Meng and Xin Liu and Zhaodan Kong and Xin Chen},
journal= {arXiv preprint arXiv:2210.13507},
year = {2023}
}