用大模型优化代码化策略，让游戏智能体自进化
Learning Game-Playing Agents with Generative Code Optimization
- 将策略写成可执行的Python代码，通过运行轨迹和语言反馈迭代优化
- 在Atari游戏上性能媲美深度强化学习，训练时间更短、环境交互少得多
- 适合想快速构建高效、可解释游戏智能体的研究者与开发者
我们提出一种生成式优化方法来学习游戏博弈智能体，其中策略以Python程序形式表示，并利用大语言模型（LLMs）进行优化。该方法将决策策略视为自我演化的代码，以当前观测为输入，输出游戏内动作，使智能体可通过执行轨迹和自然语言反馈实现自我改进，几乎无需人工干预。应用于Atari游戏时，我们的游戏策略Python程序表现达到深度强化学习基线水平，但训练时间显著缩短，环境交互次数大幅减少。本工作展示了程序化策略表示在构建高效、可适应、具备复杂长程推理能力智能体方面的潜力。
原文摘要 · Abstract (English)
We present a generative optimization approach for learning game-playing agents, where policies are represented as Python programs and refined using large language models (LLMs). Our method treats decision-making policies as self-evolving code, with current observation as input and an in-game action as output, enabling agents to self-improve through execution traces and natural language feedback with minimal human intervention. Applied to Atari games, our game-playing Python program achieves performance competitive with deep reinforcement learning (RL) baselines while using significantly less training time and much fewer environment interactions. This work highlights the promise of programmatic policy representations for building efficient, adaptable agents capable of complex, long-horizon reasoning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。