paper-with-me

Papers

Exploring Gradient Explosion in Generative Adversarial Imitation Learning: A Probabilistic Perspective

2023-12-18 · Wanying Wang, Yichen Zhu, Yirui Zhou, Chaomin Shen, Jian Tang, Zhiyuan Xu, Yaxin Peng, Yangchun Zhang

Generative Adversarial Imitation Learning (GAIL) stands as a cornerstone approach in imitation learning. This paper investigates the gradient explosion in two types of GAIL: GAIL with deterministic policy (DE-GAIL) and GAIL with stochastic policy (ST-GAIL). We begin with the observation that the training can be highly unstable for DE-GAIL at the beginning of the training phase and end up divergence. Conversely, the ST-GAIL training trajectory remains consistent, reliably converging. To shed light on these disparities, we provide an explanation from a theoretical perspective. By establishing a probabilistic lower bound for GAIL, we demonstrate that gradient explosion is an inevitable outcome for DE-GAIL due to occasionally large expert-imitator policy disparity, whereas ST-GAIL does not have the issue with it. To substantiate our assertion, we illustrate how modifications in the reward function can mitigate the gradient explosion challenge. Finally, we propose CREDO, a simple yet effective strategy that clips the reward function during the training phase, allowing the GAIL to enjoy high data efficiency and stable trainability.

📄 PDF Abstract BibTeX arXiv:2312.11214

Code (0)

등록된 구현이 없습니다.

Tasks

Imitation Learning

Methods 이 논문이 사용한 방법론

GAIL Generative Adversarial Imitation Learning presents a new general framework for directly extracting a policy from data, as if it were obtained by reinforcement learning…

Similar Papers 제목 키워드 기반

Wasserstein Distance guided Adversarial Imitation Learning with Reward Shape Exploration

2020-06-05 · Ming Zhang, Yawei Wang, Xiaoteng Ma, Li Xia 외

The generative adversarial imitation learning (GAIL) has provided an adversarial learning framework for imitating expert policy from demonstrations in high-dimensional continuous tasks. However, almost all GAIL and its e…

continuous-controlContinuous ControlImitation LearningMuJoCo

Exploring Generative Adversarial Networks for Text-to-Image Generation with Evolution Strategies

2022-07-06 · Victor Costa, Nuno Lourenço, João Correia, Penousal Machado

In the context of generative models, text-to-image generation achieved impressive results in recent years. Models using different approaches were proposed and trained in huge datasets of pairs of texts and images. Howeve…

DiversityImage GenerationText to Image GenerationText-to-Image Generation

IGAN: A New Inception-based Model for Stable and High-Fidelity Image Synthesis Using Generative Adversarial Networks

2026-01-13 · Ahmed A. Hashim, Ali Al-Shuwaili, Asraa Saeed, Ali Al-Bayaty arxiv

Generative Adversarial Networks (GANs) face a significant challenge of striking an optimal balance between high-quality image generation and training stability. Recent techniques, such as DCGAN, BigGAN, and StyleGAN, imp…

Image Generation

End-to-End Differentiable Adversarial Imitation Learning

2017-08-01 · ICML 2017 8 · Nir Baram, Oron Anschel, Itai Caspi, Shie Mannor

Generative Adversarial Networks (GANs) have been successfully applied to the problem of policy imitation in a model-free setup. However, the computation graph of GANs, that include a stochastic policy as the generat…

Imitation Learning

Model-based Adversarial Imitation Learning

2016-12-07 · Nir Baram, Oron Anschel, Shie Mannor

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 discriminate…

Imitation LearningmodelMuJoCo