Pearl: Parallel Evolutionary and Reinforcement Learning Library

TL;DR

Pearl is an open-source Python library that integrates reinforcement learning and evolutionary computation, enabling researchers to compare, combine, and visualize these approaches efficiently.

Contribution

The paper introduces Pearl, the first library to combine reinforcement learning and evolutionary algorithms for enhanced research and visualization capabilities.

Findings

Facilitates comparison between RL and evolutionary algorithms

Supports hybrid approaches combining both methods

Provides comprehensive visualization tools

Abstract

Reinforcement learning is increasingly finding success across domains where the problem can be represented as a Markov decision process. Evolutionary computation algorithms have also proven successful in this domain, exhibiting similar performance to the generally more complex reinforcement learning. Whilst there exist many open-source reinforcement learning and evolutionary computation libraries, no publicly available library combines the two approaches for enhanced comparison, cooperation, or visualization. To this end, we have created Pearl (https://github.com/LondonNode/Pearl), an open source Python library designed to allow researchers to rapidly and conveniently perform optimized reinforcement learning, evolutionary computation and combinations of the two. The key features within Pearl include: modular and expandable components, opinionated module settings, Tensorboard integration, custom callbacks and comprehensive visualizations.