Reinforcement learning algorithms are known to be sample inefficient, and often performance on one task can be substantially improved by leveraging information (e.g., via pre-training) on other related tasks. In this work, we propose a technique to achieve such knowledge transfer in cases where agent trajectories contain sensitive or private information, such as in the healthcare domain. Our approach leverages a differentially private policy evaluation algorithm to initialize an actor-critic model and improve the effectiveness of learning in downstream tasks. We empirically show this technique increases sample efficiency in resource-constrained control problems while preserving the privacy of trajectories collected in an upstream task.
@article{arxiv.1910.05876,
title = {Actor Critic with Differentially Private Critic},
author = {Jonathan Lebensold and William Hamilton and Borja Balle and Doina Precup},
journal= {arXiv preprint arXiv:1910.05876},
year = {2019}
}
Comments
6 Pages, Presented at the Privacy in Machine Learning Workshop, NeurIPS 2019