Proximal Point Imitation LearningDownload PDF

Published: 31 Oct 2022, 18:00, Last Modified: 04 Jan 2023, 17:00NeurIPS 2022 AcceptReaders: Everyone
Keywords: imitation learning theory, learning from demonstrations, proximal point method
TL;DR: A theoretically grounded new algorithm for imitation learning.
Abstract: This work develops new algorithms with rigorous efficiency guarantees for infinite horizon imitation learning (IL) with linear function approximation without restrictive coherence assumptions. We begin with the minimax formulation of the problem and then outline how to leverage classical tools from optimization, in particular, the proximal-point method (PPM) and dual smoothing, for online and offline IL, respectively. Thanks to PPM, we avoid nested policy evaluation and cost updates for online IL appearing in the prior literature. In particular, we do away with the conventional alternating updates by the optimization of a single convex and smooth objective over both cost and $Q$-functions. When solved inexactly, we relate the optimization errors to the suboptimality of the recovered policy. As an added bonus, by re-interpreting PPM as dual smoothing with the expert policy as a center point, we also obtain an offline IL algorithm enjoying theoretical guarantees in terms of required expert trajectories. Finally, we achieve convincing empirical performance for both linear and neural network function approximation.
Supplementary Material: pdf
22 Replies