papers AI Learner
The Github is limit! Click to go to the new site.

Let's Play Again: Variability of Deep Reinforcement Learning Agents in Atari Environments

2019-04-12
Kaleigh Clary, Emma Tosch, John Foley, David Jensen

Abstract

Reproducibility in reinforcement learning is challenging: uncontrolled stochasticity from many sources, such as the learning algorithm, the learned policy, and the environment itself have led researchers to report the performance of learned agents using aggregate metrics of performance over multiple random seeds for a single environment. Unfortunately, there are still pernicious sources of variability in reinforcement learning agents that make reporting common summary statistics an unsound metric for performance. Our experiments demonstrate the variability of common agents used in the popular OpenAI Baselines repository. We make the case for reporting post-training agent performance as a distribution, rather than a point estimate.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1904.06312

PDF

http://arxiv.org/pdf/1904.06312


Similar Posts

Comments