LLM4RL: Enhancing Reinforcement Learning with Large Language Models

Jiehan Zhou(Shandong University of Science and Technology), Yang Zhao(Shandong University of Science and Technology), Jiahong Liu(Shandong University of Science and Technology), Peijun Dong(Shandong University of Science and Technology), Xiaoyu Luo(Kyung Hee University), Hang Tao(Shandong University of Science and Technology), Shi‐Chung Chang(Western University), Hanjiang Luo(Shandong University of Science and Technology)
Unknown
August 6, 2024
10.1109/ccece59415.2024.10667224
Cited by 5

Abstract

Integrating large language models (LLMs) into reinforcement learning (RL) promises to enhance the learning performance. Traditional RL faces challenges in industrial settings, including complex environments, safety concerns, and multimodal data. As powerful tools for contextual learning and reasoning, LLMs can address issues inherent in traditional RL. This paper introduces a generic LLM4RL framework, and investigates how LLM4RL can improve learning performance in autonomous driving.

Related Papers

Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methods
Yuji Cao, Huan Zhao, Yuheng Cheng et al.|IEEE Transactions on Neural Networks and Learning Systems|2024|104
Guiding Pretraining in Reinforcement Learning with Large Language Models
Yuqing Du, Olivia Watkins, Zihan Wang et al.|arXiv (Cornell University)|2023|39
Exploring Bus Stop Mobility Pattern: A Multi-Pattern Deep Learning Prediction Framework
Xiangjie Kong, Zhehui Shen, Kailai Wang et al.|IEEE Transactions on Intelligent Transportation Systems|2024|30
Reward Design with Language Models
Minae Kwon, Sang Michael Xie, Kalesha Bullard et al.|arXiv (Cornell University)|2023|21
Auto MC-Reward: Automated Dense Reward Design with Large Language Models for Minecraft
Hao Li, Xue Yang, Zhaokai Wang et al.|Unknown|2024|19