A-LAMP: Agentic LLM-Based Framework for Automated MDP Modeling and Policy Generation
This addresses the problem of automating RL application for practitioners by reducing modeling errors and misalignment, though it is incremental as it builds on existing LLM-based methods.
The paper tackles the challenge of automating the conversion of natural language task descriptions into formal MDPs and trained policies in reinforcement learning, resulting in a framework that consistently achieves higher policy generation capability than a single state-of-the-art LLM model, with its lightweight variant approaching the performance of larger models.
Applying reinforcement learning (RL) to real-world tasks requires converting informal descriptions into a formal Markov decision process (MDP), implementing an executable environment, and training a policy agent. Automating this process is challenging due to modeling errors, fragile code, and misaligned objectives, which often impede policy training. We introduce an agentic large language model (LLM)-based framework for automated MDP modeling and policy generation (A-LAMP), that automatically translates free-form natural language task descriptions into an MDP formulation and trained policy. The framework decomposes modeling, coding, and training into verifiable stages, ensuring semantic alignment throughout the pipeline. Across both classic control and custom RL domains, A-LAMP consistently achieves higher policy generation capability than a single state-of-the-art LLM model. Notably, even its lightweight variant, which is built on smaller language models, approaches the performance of much larger models. Failure analysis reveals why these improvements occur. In addition, a case study also demonstrates that A-LAMP generates environments and policies that preserve the task's optimality, confirming its correctness and reliability.