Offline Reinforcement Learning with Differential Privacy

Dan Qiao, Yu-Xiang Wang

Keywords: Differential privacy, offline reinforcement learning, reinforcement learning theory

TL;DR: This paper presents the first provably efficient algorithms for offline RL with DP guarantee.

Abstract: The offline reinforcement learning (RL) problem is often motivated by the need to learn data-driven decision policies in financial, legal and healthcare applications. However, the learned policy could retain sensitive information of individuals in the training data (e.g., treatment and outcome of patients), thus susceptible to various privacy risks. We design offline RL algorithms with differential privacy guarantees which provably prevent such risks. These algorithms also enjoy strong instance-dependent learning bounds under both tabular and linear Markov Decision Process (MDP) settings. Our theory and simulation suggest that the privacy guarantee comes at (almost) no drop in utility comparing to the non-private counterpart for a medium-size dataset.

Submission Number: 2149