View a PDF of the paper titled Efficient Multi-Policy Evaluation for Reinforcement Learning, by Shuze Daniel Liu and 2 other authors
Abstract:To unbiasedly evaluate multiple target policies, the dominant approach among RL practitioners is to run and evaluate each target policy separately. However, this evaluation method is far from efficient because samples are not shared across policies, and running target policies to evaluate themselves is actually not optimal. In this paper, we address these two weaknesses by designing a tailored behavior policy to reduce the variance of estimators across all target policies. Theoretically, we prove that executing this behavior policy with manyfold fewer samples outperforms on-policy evaluation on every target policy under characterized conditions. Empirically, we show our estimator has a substantially lower variance compared with previous best methods and achieves state-of-the-art performance in a broad range of environments.
Submission history
From: Shuze Liu [view email]
[v1]
Fri, 16 Aug 2024 12:33:40 UTC (2,477 KB)
[v2]
Fri, 20 Dec 2024 05:45:00 UTC (2,449 KB)
Source link
lol