Prioritized Sequence Experience Replay

TL;DR

This paper introduces Prioritized Sequence Experience Replay (PSER), a new method that prioritizes sequences of experiences to improve learning efficiency and performance in reinforcement learning, outperforming existing prioritized experience replay methods.

Contribution

The paper proposes PSER, a novel framework for sequence prioritization in experience replay, with theoretical convergence guarantees and empirical performance improvements over PER.

Findings

PSER converges faster than PER in theory.

Empirically, PSER outperforms PER on Atari benchmarks.

PSER enhances learning efficiency in deep reinforcement learning.

Abstract

Experience replay is widely used in deep reinforcement learning algorithms and allows agents to remember and learn from experiences from the past. In an effort to learn more efficiently, researchers proposed prioritized experience replay (PER) which samples important transitions more frequently. In this paper, we propose Prioritized Sequence Experience Replay (PSER) a framework for prioritizing sequences of experience in an attempt to both learn more efficiently and to obtain better performance. We compare the performance of PER and PSER sampling techniques in a tabular Q-learning environment and in DQN on the Atari 2600 benchmark. We prove theoretically that PSER is guaranteed to converge faster than PER and empirically show PSER substantially improves upon PER.