Towards Continual Reinforcement Learning: A Review and Perspectives

TL;DR

This paper reviews the current state of continual reinforcement learning, proposing a taxonomy of formulations and approaches, discussing evaluation benchmarks, and highlighting open challenges for future research in realistic, non-stationary environments.

Contribution

It provides a comprehensive taxonomy of continual RL formulations and approaches, and discusses evaluation metrics and open problems in the field.

Findings

Taxonomy of non-stationarity in continual RL

Overview of benchmarks and evaluation metrics

Identification of open challenges and future directions

Abstract

In this article, we aim to provide a literature review of different formulations and approaches to continual reinforcement learning (RL), also known as lifelong or non-stationary RL. We begin by discussing our perspective on why RL is a natural fit for studying continual learning. We then provide a taxonomy of different continual RL formulations by mathematically characterizing two key properties of non-stationarity, namely, the scope and driver non-stationarity. This offers a unified view of various formulations. Next, we review and present a taxonomy of continual RL approaches. We go on to discuss evaluation of continual RL agents, providing an overview of benchmarks used in the literature and important metrics for understanding agent performance. Finally, we highlight open problems and challenges in bridging the gap between the current state of continual RL and findings in neuroscience. While still in its early days, the study of continual RL has the promise to develop better incremental reinforcement learners that can function in increasingly realistic applications where non-stationarity plays a vital role. These include applications such as those in the fields of healthcare, education, logistics, and robotics.