From Laws to Motivation: Guiding Exploration through Law-Based Reasoning and Rewards

TL;DR

This paper introduces a law-based reasoning approach that extracts experience from interaction records to guide exploration and improve the performance of RL and LLM agents in complex environments.

Contribution

It presents a novel method to model environment laws from experience and use them as internal motivation, enhancing agent exploration and decision-making.

Findings

Agents benefit from experience-guided exploration

Performance improvements in Crafter environment

Flexible use of language-expressed experience for reasoning and rewards

Abstract

Large Language Models (LLMs) and Reinforcement Learning (RL) are two powerful approaches for building autonomous agents. However, due to limited understanding of the game environment, agents often resort to inefficient exploration and trial-and-error, struggling to develop long-term strategies or make decisions. We propose a method that extracts experience from interaction records to model the underlying laws of the game environment, using these experience as internal motivation to guide agents. These experience, expressed in language, are highly flexible and can either assist agents in reasoning directly or be transformed into rewards for guiding training. Our evaluation results in Crafter demonstrate that both RL and LLM agents benefit from these experience, leading to improved overall performance.