Picture for Honghao Wei

Honghao Wei

Marvel: Accelerating Safe Online Reinforcement Learning with Finetuned Offline Policy

Add code
Dec 05, 2024
Viaarxiv icon

Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization

Add code
Oct 25, 2024
Figure 1 for Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization
Figure 2 for Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization
Figure 3 for Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization
Figure 4 for Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization
Viaarxiv icon

Adversarially Trained Actor Critic for offline CMDPs

Add code
Jan 01, 2024
Viaarxiv icon

Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration

Add code
Dec 22, 2023
Viaarxiv icon

Model-Free, Regret-Optimal Best Policy Identification in Online CMDPs

Add code
Sep 27, 2023
Viaarxiv icon

Sample Efficient Reinforcement Learning in Mixed Systems through Augmented Samples and Its Applications to Queueing Networks

Add code
May 25, 2023
Viaarxiv icon

Provably Efficient Model-Free Algorithms for Non-stationary CMDPs

Add code
Mar 10, 2023
Viaarxiv icon

Scalable and Sample Efficient Distributed Policy Gradient Algorithms in Multi-Agent Networked Systems

Add code
Dec 13, 2022
Viaarxiv icon

A Provably-Efficient Model-Free Algorithm for Constrained Markov Decision Processes

Add code
Jun 03, 2021
Figure 1 for A Provably-Efficient Model-Free Algorithm for Constrained Markov Decision Processes
Viaarxiv icon

FORK: A Forward-Looking Actor For Model-Free Reinforcement Learning

Add code
Oct 04, 2020
Figure 1 for FORK: A Forward-Looking Actor For Model-Free Reinforcement Learning
Figure 2 for FORK: A Forward-Looking Actor For Model-Free Reinforcement Learning
Figure 3 for FORK: A Forward-Looking Actor For Model-Free Reinforcement Learning
Figure 4 for FORK: A Forward-Looking Actor For Model-Free Reinforcement Learning
Viaarxiv icon