How to Choose a Reinforcement-Learning Algorithm

TL;DR

This paper provides a structured overview and guidelines to help practitioners select appropriate reinforcement learning algorithms and action-distribution families for specific tasks.

Contribution

It offers a comprehensive, organized framework and online interactive tool to streamline the selection process of reinforcement learning methods.

Findings

Structured overview of RL algorithms and their properties

Guidelines for choosing suitable RL methods based on task characteristics

Online interactive tool for algorithm selection

Abstract

The field of reinforcement learning offers a large variety of concepts and methods to tackle sequential decision-making problems. This variety has become so large that choosing an algorithm for a task at hand can be challenging. In this work, we streamline the process of choosing reinforcement-learning algorithms and action-distribution families. We provide a structured overview of existing methods and their properties, as well as guidelines for when to choose which methods. An interactive version of these guidelines is available online at https://rl-picker.github.io/.