Enhancing Q-Learning with Large Language Model Heuristics

arXiv:2405.03341v1 Announce Type: new
Abstract: Q-learning excels in learning from feedback within sequential decision-making tasks but requires extensive sampling for significant improvements. Although reward shaping is a powerful technique for enhancing learning efficiency, it can introduce biases that affect agent performance. Furthermore, potential-based reward shaping is constrained as it does not allow for reward modifications based on actions or terminal states, potentially limiting its effectiveness in complex environments. Additionally, large language models (LLMs) can achieve zero-shot learning, but this is …

abstract agent arxiv biases cs.ai cs.lg decision efficiency feedback heuristics improvements language language model large language large language model making performance q-learning sampling tasks type