Investigating the Interplay of Prioritized Replay and Generalization

TL;DR

This paper examines how prioritized replay influences learning in reinforcement learning, revealing its benefits and limitations with neural networks and proposing improvements for specific settings.

Contribution

It provides a detailed analysis of prioritized experience replay's effectiveness across different scenarios and introduces enhancements for its application in tabular and noisy environments.

Findings

PER improves value propagation in tabular settings

PER does not consistently outperform uniform replay with neural networks

Mitigations can reduce error spikes but often do not surpass uniform replay

Abstract

Experience replay, the reuse of past data to improve sample efficiency, is ubiquitous in reinforcement learning. Though a variety of smart sampling schemes have been introduced to improve performance, uniform sampling by far remains the most common approach. One exception is Prioritized Experience Replay (PER), where sampling is done proportionally to TD errors, inspired by the success of prioritized sweeping in dynamic programming. The original work on PER showed improvements in Atari, but follow-up results were mixed. In this paper, we investigate several variations on PER, to attempt to understand where and when PER may be useful. Our findings in prediction tasks reveal that while PER can improve value propagation in tabular settings, behavior is significantly different when combined with neural networks. Certain mitigations $-$ like delaying target network updates to control generalization and using estimates of expected TD errors in PER to avoid chasing stochasticity $-$ can avoid large spikes in error with PER and neural networks but generally do not outperform uniform replay. In control tasks, none of the prioritized variants consistently outperform uniform replay. We present new insight into the interaction between prioritization, bootstrapping, and neural networks and propose several improvements for PER in tabular settings and noisy domains.