# An Atari Model Zoo for Analyzing, Visualizing, and Comparing Deep Reinforcement Learning Agents

## TL;DR

This paper introduces the Atari Zoo framework, a collection of trained deep reinforcement learning models for Atari games, enabling easier analysis, visualization, and comparison of different algorithms' learned representations.

## Contribution

It provides a scalable, accessible platform with trained models and analysis tools, facilitating understanding and comparison of deep RL algorithms on Atari benchmarks.

## Key findings

- Initial comparisons reveal distinct performance patterns among algorithms.
- Visualization uncovers differences in learned representations.
- The framework simplifies analysis of deep RL models.

## Abstract

Much human and computational effort has aimed to improve how deep reinforcement learning algorithms perform on benchmarks such as the Atari Learning Environment. Comparatively less effort has focused on understanding what has been learned by such methods, and investigating and comparing the representations learned by different families of reinforcement learning (RL) algorithms. Sources of friction include the onerous computational requirements, and general logistical and architectural complications for running Deep RL algorithms at scale. We lessen this friction, by (1) training several algorithms at scale and releasing trained models, (2) integrating with a previous Deep RL model release, and (3) releasing code that makes it easy for anyone to load, visualize, and analyze such models. This paper introduces the Atari Zoo framework, which contains models trained across benchmark Atari games, in an easy-to-use format, as well as code that implements common modes of analysis and connects such models to a popular neural network visualization library. Further, to demonstrate the potential of this dataset and software package, we show initial quantitative and qualitative comparisons between the performance and representations of several deep RL algorithms, highlighting interesting and previously unknown distinctions between them.