Model-based Adversarial Imitation Learning

TL;DR

This paper introduces MAIL, a model-based adversarial imitation learning algorithm that uses a forward model for differentiability, reducing environment interactions and hyper-parameter tuning, and achieving superior results in MuJoCo tasks.

Contribution

MAIL is the first to incorporate a forward model into adversarial imitation learning, enabling fully differentiable training and improved efficiency.

Findings

Outperforms current state-of-the-art in MuJoCo tasks

Requires fewer environment interactions

Needs fewer hyper-parameters to tune

Abstract

Generative adversarial learning is a popular new approach to training generative models which has been proven successful for other related problems as well. The general idea is to maintain an oracle $D$ that discriminates between the expert's data distribution and that of the generative model $G$. The generative model is trained to capture the expert's distribution by maximizing the probability of $D$ misclassifying the data it generates. Overall, the system is \emph{differentiable} end-to-end and is trained using basic backpropagation. This type of learning was successfully applied to the problem of policy imitation in a model-free setup. However, a model-free approach does not allow the system to be differentiable, which requires the use of high-variance gradient estimations. In this paper we introduce the Model based Adversarial Imitation Learning (MAIL) algorithm. A model-based approach for the problem of adversarial imitation learning. We show how to use a forward model to make the system fully differentiable, which enables us to train policies using the (stochastic) gradient of $D$. Moreover, our approach requires relatively few environment interactions, and fewer hyper-parameters to tune. We test our method on the MuJoCo physics simulator and report initial results that surpass the current state-of-the-art.