Generative Adversarial Imitation

Generative Adversarial Imitation Learning (GAIL) aims to train agents to mimic expert behavior by learning an implicit reward function, avoiding the need for explicit reward engineering. Current research focuses on improving GAIL's stability and sample efficiency, often employing techniques from control theory and incorporating hierarchical models or state observers to handle complex scenarios and high-dimensional data, such as in autonomous driving and robotics. These advancements are significant for applications requiring robust and efficient policy learning from expert demonstrations, impacting fields like autonomous vehicle development, robot navigation in crowded environments, and wireless network optimization.

Papers

November 15, 2024

Off-Dynamics Reinforcement Learning via Domain Adaptation and Reward Augmented Imitation
Yihong Guo, Yixuan Wang, Yuanyuan Shi, Pan Xu, Anqi Liu
Domain Adaptation Imitation Learning Policy OpTimization Reward Report Learning Augmented Generative Adversarial Imitation

September 27, 2024

Enhancing Spectrum Efficiency in 6G Satellite Networks: A GAIL-Powered Policy Learning via Asynchronous Federated Inverse Reinforcement Learning
Sheikh Salman Hassan, Yu Min Park, Yan Kyaw Tun, Walid Saad, Zhu Han, Choong Seon Hong
Gene Level GNN Policy Learning Federated Reinforcement Learning Traditional Reinforcement Learning Generative Adversarial Imitation Wireless Network Optimization Spectrum Efficiency

August 7, 2024

Navigating the Human Maze: Real-Time Robot Pathfinding with Generative Imitation Learning
Martin Moder, Stephen Adhisaputra, Josef Pauli
Robot Trajectory Maze Environment Goal Conditioned Sampling Based Model Predictive Control Generative Adversarial Imitation Proactive Robot

August 6, 2024

Adversarial Safety-Critical Scenario Generation using Naturalistic Human Driving Priors
Kunkun Hao, Yonggang Luo, Wen Cui, Yuqiao Bai, Jucheng Yang, Songyang Yan, Yuxi Pan, Zijiang Yang
Autonomous Vehicle Adversarial Scenario Naturalistic Driving Intelligent Driver Model Generative Adversarial Imitation

February 26, 2024

C-GAIL: Stabilizing Generative Adversarial Imitation Learning with Control Theory
Tianjiao Luo, Tim Pearce, Huayu Chen, Jianfei Chen, Jun Zhu
GAN Model Policy Reinforcement Learning Generative Training Generative Adversarial Imitation Learning Control Theory Generative Adversarial Imitation

April 4, 2023

Quantum Imitation Learning
Zhihao Cheng, Kaining Zhang, Li Shen, Dacheng Tao
Inverse Reinforcement Learning Variational Quantum Circuit Quantum Advantage Imitation Game Generative Adversarial Imitation

October 18, 2022

Hierarchical Model-Based Imitation Learning for Planning in Autonomous Driving
Eli Bronstein, Mark Palatucci, Dominik Notz, Brandyn White, Alex Kuefler, Yiren Lu, Supratik Paul, Payam Nikdel, Paul Mougin, Hongge Chen, Justin Fu, Austin Abrams, Punit Shah, Evan Racah, Benjamin Frenkel, Shimon Whiteson, Dragomir Anguelov
Autonomous Driving Imitation Learning Task Planning Driving Scenario Expert Trajectory Urban Autonomous Driving Generative Adversarial Imitation Contrastive Imitation

April 5, 2022

SHAIL: Safety-Aware Hierarchical Adversarial Imitation Learning for Autonomous Driving in Urban Environments
Arec Jamgochian, Etienne Buehrle, Johannes Fischer, Mykel J. Kochenderfer
Autonomous Driving Urban Environment Adversarial Imitation Learning Autonomous Driving Policy Generative Adversarial Imitation

February 1, 2022

Adversarial Imitation Learning from Video using a State Observer
Haresh Karnan, Garrett Warnell, Faraz Torabi, Peter Stone
Source Video Adversarial Imitation Learning Adversarial Imitation State Observation Learning Based Observer Generative Adversarial Imitation Generative Visual

Generative Adversarial Imitation

Papers

Off-Dynamics Reinforcement Learning via Domain Adaptation and Reward Augmented Imitation

Enhancing Spectrum Efficiency in 6G Satellite Networks: A GAIL-Powered Policy Learning via Asynchronous Federated Inverse Reinforcement Learning

Navigating the Human Maze: Real-Time Robot Pathfinding with Generative Imitation Learning

Adversarial Safety-Critical Scenario Generation using Naturalistic Human Driving Priors

C-GAIL: Stabilizing Generative Adversarial Imitation Learning with Control Theory

Quantum Imitation Learning

Hierarchical Model-Based Imitation Learning for Planning in Autonomous Driving

SHAIL: Safety-Aware Hierarchical Adversarial Imitation Learning for Autonomous Driving in Urban Environments

Adversarial Imitation Learning from Video using a State Observer