Reward Guidance for Reinforcement Learning Tasks Based on Large Language Models: The LMGT Framework

Yongxin Deng, Xihe Qiu, Jue Chen +1

The inherent uncertainty in the environmental transition model of Reinforcement Learning (RL) necessitates a delicate balance between exploration and exploitation. This balance is…