Vizarel: A System to Help Better Understand RL Agents

TL;DR

Vizarel is a proposed visualization system designed to enhance understanding and interpretability of reinforcement learning agents, inspired by tools used in supervised learning but tailored for RL's unique challenges.

Contribution

The paper introduces the initial design and features of Vizarel, a system aimed at improving interpretability of RL agents through visualization.

Findings

Identified key features needed for RL interpretability tools.

Proposed a prototype platform for experimenting with RL interpretability.

Laid groundwork for future development of RL visualization tools.

Abstract

Visualization tools for supervised learning have allowed users to interpret, introspect, and gain intuition for the successes and failures of their models. While reinforcement learning practitioners ask many of the same questions, existing tools are not applicable to the RL setting. In this work, we describe our initial attempt at constructing a prototype of these ideas, through identifying possible features that such a system should encapsulate. Our design is motivated by envisioning the system to be a platform on which to experiment with interpretable reinforcement learning.