Saved in:
Bibliographic Details
Main Authors: Zhang, Kaifeng, Zhao, Rui, Zhang, Ziming, Gao, Yang
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2206.11004
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910314866409472
author Zhang, Kaifeng
Zhao, Rui
Zhang, Ziming
Gao, Yang
author_facet Zhang, Kaifeng
Zhao, Rui
Zhang, Ziming
Gao, Yang
contents Reinforcement learning (RL) provides a powerful framework for decision-making, but its application in practice often requires a carefully designed reward function. Adversarial Imitation Learning (AIL) sheds light on automatic policy acquisition without access to the reward signal from the environment. In this work, we propose Auto-Encoding Adversarial Imitation Learning (AEAIL), a robust and scalable AIL framework. To induce expert policies from demonstrations, AEAIL utilizes the reconstruction error of an auto-encoder as a reward signal, which provides more information for optimizing policies than the prior discriminator-based ones. Subsequently, we use the derived objective functions to train the auto-encoder and the agent policy. Experiments show that our AEAIL performs superior compared to state-of-the-art methods on both state and image based environments. More importantly, AEAIL shows much better robustness when the expert demonstrations are noisy.
format Preprint
id arxiv_https___arxiv_org_abs_2206_11004
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Auto-Encoding Adversarial Imitation Learning
Zhang, Kaifeng
Zhao, Rui
Zhang, Ziming
Gao, Yang
Machine Learning
Reinforcement learning (RL) provides a powerful framework for decision-making, but its application in practice often requires a carefully designed reward function. Adversarial Imitation Learning (AIL) sheds light on automatic policy acquisition without access to the reward signal from the environment. In this work, we propose Auto-Encoding Adversarial Imitation Learning (AEAIL), a robust and scalable AIL framework. To induce expert policies from demonstrations, AEAIL utilizes the reconstruction error of an auto-encoder as a reward signal, which provides more information for optimizing policies than the prior discriminator-based ones. Subsequently, we use the derived objective functions to train the auto-encoder and the agent policy. Experiments show that our AEAIL performs superior compared to state-of-the-art methods on both state and image based environments. More importantly, AEAIL shows much better robustness when the expert demonstrations are noisy.
title Auto-Encoding Adversarial Imitation Learning
topic Machine Learning
url https://arxiv.org/abs/2206.11004