An empirical investigation of the challenges of real-world reinforcement learning

TL;DR

This paper identifies and formalizes key challenges hindering the deployment of reinforcement learning in real-world systems, analyzing their impact on algorithms and proposing an open-source benchmark suite.

Contribution

It introduces a formal framework for real-world RL challenges, analyzes their effects on algorithms, and provides an open-source benchmark suite for future research.

Findings

Challenges significantly affect RL algorithm performance

Existing solutions partially address some challenges

The realworldrl-suite enables standardized evaluation

Abstract

Reinforcement learning (RL) has proven its worth in a series of artificial domains, and is beginning to show some successes in real-world scenarios. However, much of the research advances in RL are hard to leverage in real-world systems due to a series of assumptions that are rarely satisfied in practice. In this work, we identify and formalize a series of independent challenges that embody the difficulties that must be addressed for RL to be commonly deployed in real-world systems. For each challenge, we define it formally in the context of a Markov Decision Process, analyze the effects of the challenge on state-of-the-art learning algorithms, and present some existing attempts at tackling it. We believe that an approach that addresses our set of proposed challenges would be readily deployable in a large number of real world problems. Our proposed challenges are implemented in a suite of continuous control environments called the realworldrl-suite which we propose an as an open-source benchmark.