Saved in:
Bibliographic Details
Main Authors: Tan, Heng, Yan, Hua, Yang, Yu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.20671
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908380675702784
author Tan, Heng
Yan, Hua
Yang, Yu
author_facet Tan, Heng
Yan, Hua
Yang, Yu
contents While reinforcement learning (RL) has achieved notable success in various domains, training effective policies for complex tasks remains challenging. Agents often converge to local optima and fail to maximize long-term rewards. Existing approaches to mitigate training bottlenecks typically fall into two categories: (i) Automated policy refinement, which identifies critical states from past trajectories to guide policy updates, but suffers from costly and uncertain model training; and (ii) Human-in-the-loop refinement, where human feedback is used to correct agent behavior, but this does not scale well to environments with large or continuous action spaces. In this work, we design a large language model-guided policy modulation framework that leverages LLMs to improve RL training without additional model training or human intervention. We first prompt an LLM to identify critical states from a sub-optimal agent's trajectories. Based on these states, the LLM then provides action suggestions and assigns implicit rewards to guide policy refinement. Experiments across standard RL benchmarks demonstrate that our method outperforms state-of-the-art baselines, highlighting the effectiveness of LLM-based explanations in addressing RL training bottlenecks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20671
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLM-Guided Reinforcement Learning: Addressing Training Bottlenecks through Policy Modulation
Tan, Heng
Yan, Hua
Yang, Yu
Artificial Intelligence
Machine Learning
While reinforcement learning (RL) has achieved notable success in various domains, training effective policies for complex tasks remains challenging. Agents often converge to local optima and fail to maximize long-term rewards. Existing approaches to mitigate training bottlenecks typically fall into two categories: (i) Automated policy refinement, which identifies critical states from past trajectories to guide policy updates, but suffers from costly and uncertain model training; and (ii) Human-in-the-loop refinement, where human feedback is used to correct agent behavior, but this does not scale well to environments with large or continuous action spaces. In this work, we design a large language model-guided policy modulation framework that leverages LLMs to improve RL training without additional model training or human intervention. We first prompt an LLM to identify critical states from a sub-optimal agent's trajectories. Based on these states, the LLM then provides action suggestions and assigns implicit rewards to guide policy refinement. Experiments across standard RL benchmarks demonstrate that our method outperforms state-of-the-art baselines, highlighting the effectiveness of LLM-based explanations in addressing RL training bottlenecks.
title LLM-Guided Reinforcement Learning: Addressing Training Bottlenecks through Policy Modulation
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2505.20671