Jian Li

Near-Optimal Online Deployment and Routing for Streaming LLMs

All-Task Convergence and Backward Transfer in Federated Domain-Incremental Learning with Partial Participation

PID-Guided Partial Alignment for Multimodal Decentralized Federated Learning

Constrained Feedback Learning for Non-Stationary Multi-Armed Bandits

Provably Efficient Exploration in Inverse Constrained Reinforcement Learning

On the Linear Speedup of Personalized Federated Reinforcement Learning with Shared Representations

DOPL: Direct Online Preference Learning for Restless Bandits with Preference Feedback

Understanding Constraint Inference in Safety-Critical Inverse Reinforcement Learning

Strategic Exploration for Inverse Constraint Inference with Efficiency Guarantee

Revisiting Critical Learning Periods in Deep Neural Networks