Evolving simple programs for playing Atari games

TL;DR

This paper demonstrates that evolving simple programs using Cartesian Genetic Programming with matrix operation functions can produce competitive Atari game controllers with less training time and interpretable strategies.

Contribution

It introduces a CGP-based approach for Atari game playing that emphasizes simplicity and interpretability, achieving competitive performance with reduced training.

Findings

Evolved programs are small yet competitive with state-of-the-art methods.

Controllers require less training time than traditional deep learning approaches.

Analysis reveals simple, effective strategies in the best evolved programs.

Abstract

Cartesian Genetic Programming (CGP) has previously shown capabilities in image processing tasks by evolving programs with a function set specialized for computer vision. A similar approach can be applied to Atari playing. Programs are evolved using mixed type CGP with a function set suited for matrix operations, including image processing, but allowing for controller behavior to emerge. While the programs are relatively small, many controllers are competitive with state of the art methods for the Atari benchmark set and require less training time. By evaluating the programs of the best evolved individuals, simple but effective strategies can be found.