Explaining Deep Reinforcement Learning Agents In The Atari Domain through a Surrogate Model
This addresses the problem of interpretability for researchers and practitioners using deep RL, though it is incremental as it builds on existing surrogate model approaches.
The paper tackles the lack of explainability in deep reinforcement learning by proposing a method to derive explanations for agents in the Atari domain, using a surrogate model that accurately approximates the agent's decision-making.
One major barrier to applications of deep Reinforcement Learning (RL) both inside and outside of games is the lack of explainability. In this paper, we describe a lightweight and effective method to derive explanations for deep RL agents, which we evaluate in the Atari domain. Our method relies on a transformation of the pixel-based input of the RL agent to an interpretable, percept-like input representation. We then train a surrogate model, which is itself interpretable, to replicate the behavior of the target, deep RL agent. Our experiments demonstrate that we can learn an effective surrogate that accurately approximates the underlying decision making of a target agent on a suite of Atari games.