Deep Reinforcement Learning That Matters

TLDR

Deep reinforcement learning has advanced many domains, yet reproducibility is hampered by non‑determinism, intrinsic variance, and insufficient significance metrics, making it hard to assess genuine progress. This study examines reproducibility challenges, experimental practices, and reporting standards in deep RL to encourage more reliable progress. By comparing metrics against common baselines, the authors demonstrate variability and propose guidelines to improve reproducibility.

Abstract

In recent years, significant progress has been made in solving challenging problems across various domains using deep reinforcement learning (RL). Reproducing existing work and accurately judging the improvements offered by novel methods is vital to sustaining this progress. Unfortunately, reproducing results for state-of-the-art deep RL methods is seldom straightforward. In particular, non-determinism in standard benchmark environments, combined with variance intrinsic to the methods, can make reported results tough to interpret. Without significance metrics and tighter standardization of experimental reporting, it is difficult to determine whether improvements over the prior state-of-the-art are meaningful. In this paper, we investigate challenges posed by reproducibility, proper experimental techniques, and reporting procedures. We illustrate the variability in reported metrics and results when comparing against common baselines and suggest guidelines to make future results in deep RL more reproducible. We aim to spur discussion about how to ensure continued progress in the field by minimizing wasted effort stemming from results that are non-reproducible and easily misinterpreted.