paper-with-me

홈 › Papers

Discriminator-Weighted Offline Imitation Learning from Suboptimal Demonstrations

2022-07-20 · Haoran Xu, Xianyuan Zhan, Honglei Yin, Huiling Qin

We study the problem of offline Imitation Learning (IL) where an agent aims to learn an optimal expert behavior policy without additional online environment interactions. Instead, the agent is provided with a supplementary offline dataset from suboptimal behaviors. Prior works that address this problem either require that expert data occupies the majority proportion of the offline dataset, or need to learn a reward function and perform offline reinforcement learning (RL) afterwards. In this paper, we aim to address the problem without additional steps of reward learning and offline RL training for the case when demonstrations contain a large proportion of suboptimal data. Built upon behavioral cloning (BC), we introduce an additional discriminator to distinguish expert and non-expert data. We propose a cooperation framework to boost the learning of both tasks, Based on this framework, we design a new IL algorithm, where the outputs of discriminator serve as the weights of the BC loss. Experimental results show that our proposed algorithm achieves higher returns and faster training speed compared to baseline algorithms.

📄 PDF Abstract BibTeX arXiv:2207.10050

Code (2)

ryanxhr/dwbc 공식 구현 pytorch
zexusun/oilca-neurips23 pytorch

Tasks

Imitation LearningOffline RLReinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…

Similar Papers 제목 키워드 기반

Discriminator-Guided Model-Based Offline Imitation Learning

2022-07-01 · Wenjia Zhang, Haoran Xu, Haoyi Niu, Peng Cheng 외

Offline imitation learning (IL) is a powerful method to solve decision-making problems from expert demonstrations without reward labels. Existing offline IL methods suffer from severe performance degeneration under limit…

Decision MakingImitation Learningmodel

An Optimal Discriminator Weighted Imitation Perspective for Reinforcement Learning

2025-04-17 · Haoran Xu, Shuozhe Li, Harshit Sikchi, Scott Niekum 외

We introduce Iterative Dual Reinforcement Learning (IDRL), a new method that takes an optimal discriminator-weighted imitation view of solving RL. Our method is motivated by a simple experiment in which we find training …

D4RLreinforcement-learningReinforcement Learning

Language-Critique Imitation Learning from Suboptimal Demonstrations

2026-07-01 · Chih-Han Yang, Dai-Jie Wu, Yun-Ping Huang, Ping-Chun Hsieh 외 arxiv

Prior work on imitation learning from suboptimal demonstrations typically relies on compressed supervision signals such as confidence estimates, discriminator scores, or importance weights. These scalar signals are inher…

Reinforcement LearningContinuous Control

Using Non-Expert Data to Robustify Imitation Learning via Offline Reinforcement Learning

2025-10-22 · Kevin Huang, Rosario Scalise, Cleah Winston, Ayush Agrawal 외 arxiv

Imitation learning has proven effective for training robots to perform complex tasks from expert human demonstrations. However, it remains limited by its reliance on high-quality, task-specific data, restricting adaptabi…

Reinforcement LearningOffline RL

Imitation Learning from Suboptimal Demonstrations via Meta-Learning An Action Ranker

2024-12-28 · Jiangdong Fan, Hongcai He, Paul Weng, Hui Xu 외

A major bottleneck in imitation learning is the requirement of a large number of expert demonstrations, which can be expensive or inaccessible. Learning from supplementary demonstrations without strict quality requiremen…

Imitation LearningMeta-Learning