PhD student, South China University of Technology
1 paper at NeurIPS 2025
We propose Continual Knowledge Adaptation for Reinforcement Learning (CKA-RL), which enables the accumulation and effective utilization of historical knowledge.
