Planning by Prioritized Sweeping with Small Backups

TL;DR

This paper introduces a novel planning backup method called small backup, which reduces computation time by focusing on individual successor states, enabling more efficient and flexible model-based reinforcement learning.

Contribution

The paper presents a new small backup operation for prioritized sweeping that decouples computation time from the number of successor states, improving planning efficiency.

Findings

Small backups enable faster planning iterations.

Prioritized sweeping with small backups outperforms traditional methods.

Enhanced flexibility leads to better planning performance.

Abstract

Efficient planning plays a crucial role in model-based reinforcement learning. Traditionally, the main planning operation is a full backup based on the current estimates of the successor states. Consequently, its computation time is proportional to the number of successor states. In this paper, we introduce a new planning backup that uses only the current value of a single successor state and has a computation time independent of the number of successor states. This new backup, which we call a small backup, opens the door to a new class of model-based reinforcement learning methods that exhibit much finer control over their planning process than traditional methods. We empirically demonstrate that this increased flexibility allows for more efficient planning by showing that an implementation of prioritized sweeping based on small backups achieves a substantial performance improvement over classical implementations.