Deep Adaptive Multi-Intention Inverse Reinforcement Learning

TL;DR

This paper introduces a deep IRL framework that adaptively learns multiple nonlinear reward functions from unlabeled demonstrations, effectively modeling multi-intention behaviors with unknown counts using Dirichlet processes.

Contribution

It proposes a novel adaptive deep IRL method employing Dirichlet processes to infer both the number of intentions and their reward functions from unlabeled data.

Findings

Outperforms existing IRL methods on benchmark tasks.

Effectively infers the number of intentions online.

Demonstrates advantages in modeling multi-intention behaviors.

Abstract

This paper presents a deep Inverse Reinforcement Learning (IRL) framework that can learn an a priori unknown number of nonlinear reward functions from unlabeled experts' demonstrations. For this purpose, we employ the tools from Dirichlet processes and propose an adaptive approach to simultaneously account for both complex and unknown number of reward functions. Using the conditional maximum entropy principle, we model the experts' multi-intention behaviors as a mixture of latent intention distributions and derive two algorithms to estimate the parameters of the deep reward network along with the number of experts' intentions from unlabeled demonstrations. The proposed algorithms are evaluated on three benchmarks, two of which have been specifically extended in this study for multi-intention IRL, and compared with well-known baselines. We demonstrate through several experiments the advantages of our algorithms over the existing approaches and the benefits of online inferring, rather than fixing beforehand, the number of expert's intentions.