Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gap

TL;DR

This paper introduces a game-theoretic framework for imitation learning based on moment matching, providing unified analysis, performance bounds, and new algorithms with strong guarantees and empirical success.

Contribution

It unifies various imitation learning algorithms through moment matching, introduces the concept of moment recoverability, and proposes three novel algorithms with theoretical and empirical advantages.

Findings

Derived bounds on policy performance for different classes of algorithms.

Introduced three new algorithms: AdVIL, AdRIL, and DAeQuIL.

Demonstrated competitive empirical performance of the proposed algorithms.

Abstract

We provide a unifying view of a large family of previous imitation learning algorithms through the lens of moment matching. At its core, our classification scheme is based on whether the learner attempts to match (1) reward or (2) action-value moments of the expert's behavior, with each option leading to differing algorithmic approaches. By considering adversarially chosen divergences between learner and expert behavior, we are able to derive bounds on policy performance that apply for all algorithms in each of these classes, the first to our knowledge. We also introduce the notion of moment recoverability, implicit in many previous analyses of imitation learning, which allows us to cleanly delineate how well each algorithmic family is able to mitigate compounding errors. We derive three novel algorithm templates (AdVIL, AdRIL, and DAeQuIL) with strong guarantees, simple implementation, and competitive empirical performance.