Multiplayer Support for the Arcade Learning Environment

TL;DR

This paper extends the Arcade Learning Environment to support multiplayer Atari games, providing a new interface integrated with PettingZoo and baseline experiments for these environments.

Contribution

It introduces a multiplayer extension to ALE, enabling reinforcement learning research in multiplayer Atari environments with a user-friendly Python interface.

Findings

Extended ALE to support multiplayer games

Integrated with PettingZoo for ease of use

Provided experimental baselines for new environments

Abstract

The Arcade Learning Environment ("ALE") is a widely used library in the reinforcement learning community that allows easy programmatic interfacing with Atari 2600 games, via the Stella emulator. We introduce a publicly available extension to the ALE that extends its support to multiplayer games and game modes. This interface is additionally integrated with PettingZoo to allow for a simple Gym-like interface in Python to interact with these games. We additionally introduce experimental baselines for all environments included.