Efficient, Low-Regret, Online Reinforcement Learning for Linear MDPs

TL;DR

This paper introduces modified online reinforcement learning algorithms for linear MDPs that reduce space and time complexity while maintaining low regret, validated through experiments on synthetic and real data.

Contribution

It proposes two variants of LSVI-UCB that alternate learning periods to improve efficiency without sacrificing regret guarantees.

Findings

Achieve low space and time complexity in experiments

Maintain sublinear regret with the modifications

Perform well on both synthetic and real-world benchmarks

Abstract

Reinforcement learning algorithms are usually stated without theoretical guarantees regarding their performance. Recently, Jin, Yang, Wang, and Jordan (COLT 2020) showed a polynomial-time reinforcement learning algorithm (namely, LSVI-UCB) for the setting of linear Markov decision processes, and provided theoretical guarantees regarding its running time and regret. In real-world scenarios, however, the space usage of this algorithm can be prohibitive due to a utilized linear regression step. We propose and analyze two modifications of LSVI-UCB, which alternate periods of learning and not-learning, to reduce space and time usage while maintaining sublinear regret. We show experimentally, on synthetic data and real-world benchmarks, that our algorithms achieve low space usage and running time, while not significantly sacrificing regret.