Year: 2021
Type: preprint
Abstract: We introduce quantile filtered imitation learning (QFIL), a novel policy improvement operator designed for offline reinforcement learning. QFIL performs policy improvement by running imitation learning on a filtered version of the offline dataset. The filtering process removes $ s,a $ pairs whose estimated Q values fall below a given quantile of the pushforward distribution over values induced by sampling actions from the behavior policy. The definitions of both the pushforward Q distribution and resulting value function quantile are key contributions of our method. We prove that QFIL gives us a safe policy improvement step with function approximation and that the choice of quantile provides a natural hyperparameter to trade off bias and variance of the improvement step. Empirically, we perform a synthetic experiment illustrating how QFIL effectively makes a bias-variance tradeoff and we see that QFIL performs well on the D4RL benchmark.
Source: arXiv (Cornell University)
Authors: Brandfonbrener, David, Whitney, William F., Ranganath, Rajesh, Bruna, Joan
Institution: New York University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3217328091