Confidence-Aware Imitation Learning from Demonstrations with Varying Optimality

TL;DR

This paper introduces Confidence-Aware Imitation Learning (CAIL), a framework that learns effective policies from demonstrations with varying optimality by jointly estimating confidence scores and policy performance, outperforming existing methods.

Contribution

The paper proposes a novel CAIL framework that jointly learns confidence scores and policies from non-optimal demonstrations, with theoretical guarantees and superior empirical results.

Findings

CAIL outperforms existing imitation learning methods in simulated and real robot experiments.

CAIL can learn successful policies even without access to optimal demonstrations.

Theoretical guarantees ensure convergence of the proposed framework.

Abstract

Most existing imitation learning approaches assume the demonstrations are drawn from experts who are optimal, but relaxing this assumption enables us to use a wider range of data. Standard imitation learning may learn a suboptimal policy from demonstrations with varying optimality. Prior works use confidence scores or rankings to capture beneficial information from demonstrations with varying optimality, but they suffer from many limitations, e.g., manually annotated confidence scores or high average optimality of demonstrations. In this paper, we propose a general framework to learn from demonstrations with varying optimality that jointly learns the confidence score and a well-performing policy. Our approach, Confidence-Aware Imitation Learning (CAIL) learns a well-performing policy from confidence-reweighted demonstrations, while using an outer loss to track the performance of our model and to learn the confidence. We provide theoretical guarantees on the convergence of CAIL and evaluate its performance in both simulated and real robot experiments. Our results show that CAIL significantly outperforms other imitation learning methods from demonstrations with varying optimality. We further show that even without access to any optimal demonstrations, CAIL can still learn a successful policy, and outperforms prior work.