Advantage Actor-Critic with Reasoner: Explaining the Agent's Behavior from an Exploratory Perspective
Muzhe Guo, Feixu Yu, Tian Lan, Fang Jin

TL;DR
This paper introduces A2CR, an interpretable reinforcement learning framework that enhances transparency by classifying and explaining agent actions, demonstrated through improved interpretability and purpose-based saliency in game environments.
Contribution
The paper presents a novel A2CR model integrating a Reasoner network into Actor-Critic RL, enabling automatic interpretation of agent behavior and purpose classification.
Findings
Reasoner label proportions vary with exploration levels in games
Purpose-based saliencies are more focused and understandable
A2CR improves interpretability of RL agents in complex environments
Abstract
Reinforcement learning (RL) is a powerful tool for solving complex decision-making problems, but its lack of transparency and interpretability has been a major challenge in domains where decisions have significant real-world consequences. In this paper, we propose a novel Advantage Actor-Critic with Reasoner (A2CR), which can be easily applied to Actor-Critic-based RL models and make them interpretable. A2CR consists of three interconnected networks: the Policy Network, the Value Network, and the Reasoner Network. By predefining and classifying the underlying purpose of the actor's actions, A2CR automatically generates a more comprehensive and interpretable paradigm for understanding the agent's decision-making process. It offers a range of functionalities such as purpose-based saliency, early failure detection, and model supervision, thereby promoting responsible and trustworthy RL.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSports Analytics and Performance · Explainable Artificial Intelligence (XAI)
