Not Logged In

Importance Resampling for Off-policy Prediction

Importance sampling (IS) is a common reweighting strategy for off-policy prediction in reinforcement learning. While it is consistent and unbiased, it can result in high variance updates to the weights for the value function. In this work, we explore a resampling strategy as an alternative to reweighting. We propose Importance Resampling (IR) for off-policy prediction, which resamples experience from a replay buffer and applies standard on-policy updates. The approach avoids using importance sampling ratios in the update, instead correcting the distribution before the update. We characterize the bias and consistency of IR, particularly compared to Weighted IS (WIS). We demonstrate in several microworlds that IR has improved sample efficiency and lower variance updates, as compared to IS and several variance-reduced IS strategies, including variants of WIS and V-trace which clips IS ratios. We also provide a demonstration showing IR improves over IS for learning a value function from images in a racing car simulator.

Citation

M. Schlegel, W. Chung, D. Graves, J. Qian, M. White. "Importance Resampling for Off-policy Prediction". Neural Information Processing Systems (NIPS), (ed: Hanna M. Wallach, Hugo Larochelle, Alina Beygelzimer, Florence d'Alche-Buc, Emily B. Fox, Roman Garnett), pp 1797-1807, December 2019.

Keywords:  
Category: In Conference
Web Links: NeurIPS

BibTeX

@incollection{Schlegel+al:NIPS19,
  author = {Matthew Schlegel and Wesley Chung and Daniel Graves and Jian Qian
    and Martha White},
  title = {Importance Resampling for Off-policy Prediction},
  Editor = {Hanna M. Wallach, Hugo Larochelle, Alina Beygelzimer, Florence
    d'Alche-Buc, Emily B. Fox, Roman Garnett},
  Pages = {1797-1807},
  booktitle = {Neural Information Processing Systems (NIPS)},
  year = 2019,
}

Last Updated: March 24, 2020
Submitted by Sabina P

University of Alberta Logo AICML Logo