Adversarial Active Exploration for Inverse Dynamics Model Learning

TL;DR

This paper introduces an adversarial active exploration method where a reinforcement learning agent and an inverse dynamics model compete, leading to improved model learning without human intervention, especially in robotic manipulation tasks.

Contribution

The paper proposes a novel adversarial framework for inverse dynamics learning that encourages exploration of challenging samples without human input, outperforming baseline methods.

Findings

Method achieves comparable performance to expert demonstration training.

Outperforms baseline models in robotic manipulation tasks.

Effective in generating moderately hard training samples.

Abstract

We present an adversarial active exploration for inverse dynamics model learning, a simple yet effective learning scheme that incentivizes exploration in an environment without any human intervention. Our framework consists of a deep reinforcement learning (DRL) agent and an inverse dynamics model contesting with each other. The former collects training samples for the latter, with an objective to maximize the error of the latter. The latter is trained with samples collected by the former, and generates rewards for the former when it fails to predict the actual action taken by the former. In such a competitive setting, the DRL agent learns to generate samples that the inverse dynamics model fails to predict correctly, while the inverse dynamics model learns to adapt to the challenging samples. We further propose a reward structure that ensures the DRL agent to collect only moderately hard samples but not overly hard ones that prevent the inverse model from predicting effectively. We evaluate the effectiveness of our method on several robotic arm and hand manipulation tasks against multiple baseline models. Experimental results show that our method is comparable to those directly trained with expert demonstrations, and superior to the other baselines even without any human priors.