Ping-Chun Hsieh

A Reward-Free Viewpoint on Multi-Objective Reinforcement Learning

Test-Time Alignment for Large Language Models via Textual Model Predictive Control

Cross-Domain Policy Optimization via Bellman Consistency and Hybrid Critics

Quality and Diversity Optimization Even From Offline Homogeneous Dataset

Unconstrained Models as Constrained Problem Solvers: Duality-Driven Adaptation without Retraining

Learning Human-Like RL Agents Through Trajectory Optimization With Action Quantization

Efficient Action-Constrained Reinforcement Learning via Acceptance-Rejection Method and Augmented MDPs

BOFormer: Learning to Solve Multi-Objective Bayesian Optimization via Non-Markovian RL

Learn out of the box: optimizing both diversity and performance in Offline Reinforcement Learning

Cross-Domain Reinforcement Learning Under Distinct State-Action Spaces Via Hybrid Q Functions

Action-Constrained Imitation Learning

Action-Constrained Imitation Learning

Cross-Domain Reinforcement Learning via Preference Consistency

Diminishing Exploration: A Minimalist Approach to Piecewise Stationary Multi-Armed Bandits