通过大语言模型引导搜索合成可解释的控制策略
Synthesizing Interpretable Control Policies through Large Language Model Guided Search
AI总结：
本文提出通过大语言模型引导搜索生成可解释的动态系统控制策略，利用标准编程语言提高透明性和可解释性，并展示在摆动单摆和杯中球任务中的应用。
摘要：
大语言模型（LLMs）、系统性评估和进化算法的结合已在组合优化和科学发现中实现了突破。我们提出将这种强大的组合扩展到动态系统的控制中，生成能够执行复杂行为的可解释控制策略。我们的新方法将控制策略表示为标准语言（如Python）中的程序。我们通过模拟评估候选控制器，并使用预训练的LLM进行进化。与传统基于学习的控制技术不同，后者依赖于黑箱神经网络来编码控制策略，我们的方法增强了透明性和可解释性。我们仍然利用大AI模型的强大功能，但仅在策略设计阶段，确保系统所有组件在运行时都保持可解释和易于验证。此外，使用标准编程语言使人类能够根据其专业知识和直觉轻松调整或适应控制器。我们通过其在摆动单摆和杯中球任务中合成可解释控制策略的应用来展示我们的方法。我们将在https://github.com/muellerlab/synthesizing_interpretable_control_policies.git上发布代码。
英文摘要：
The combination of Large Language Models (LLMs), systematic evaluation, and evolutionary algorithms has enabled breakthroughs in combinatorial optimization and scientific discovery. We propose to extend this powerful combination to the control of dynamical systems, generating interpretable control policies capable of complex behaviors. With our novel method, we represent control policies as programs in standard languages like Python. We evaluate candidate controllers in simulation and evolve them using a pre-trained LLM. Unlike conventional learning-based control techniques, which rely on black-box neural networks to encode control policies, our approach enhances transparency and interpretability. We still take advantage of the power of large AI models, but only at the policy design phase, ensuring that all system components remain interpretable and easily verifiable at runtime. Additionally, the use of standard programming languages makes it straightforward for humans to finetune or adapt the controllers based on their expertise and intuition. We illustrate our method through its application to the synthesis of an interpretable control policy for the \textit{pendulum swing-up} and the \textit{ball in cup} tasks. We make the code available at https://github.com/muellerlab/synthesizing_interpretable_control_policies.git.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
