arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2502.14693cs.CL
收藏

I-MCTS:通过反思蒙特卡洛树搜索增强代理自动机器学习

I-MCTS: Enhancing Agentic AutoML via Introspective Monte Carlo Tree Search

Ant Group(蚂蚁集团);Rutgers University(罗格斯大学)

Zujie Liang, Feng Wei, Wujiang Xu, Lin Chen, Yuxi Qian, Xinhui Wu

AI总结:

I-MCTS通过反思蒙特卡洛树搜索提升代理AutoML性能,实现4%的绝对性能提升。

摘要:

近年来,大语言模型(LLMs)在自动化机器学习任务中的进展显示出巨大的潜力。然而,现有的基于LLM的代理往往在代码生成的低多样性和次优性方面存在困难。尽管最近的研究引入了蒙特卡洛树搜索(MCTS)来解决这些问题,但在生成的思维质量和节点选择的标量反馈机制方面仍然存在局限。在本研究中,我们引入了反思蒙特卡洛树搜索(I-MCTS),一种新的方法,通过反思过程迭代扩展树节点,仔细分析父节点和兄弟节点的解决方案和结果。这有助于在搜索树中持续优化节点,从而提高整体决策过程。此外,我们集成了基于大语言模型(LLM)的值模型,以在进行全面计算展开之前直接评估每个节点的解决方案。实现了一种混合奖励机制,以无缝地将Q值从LLM估计的分数转换为实际性能分数。这使得高质量的节点可以更早地被遍历。应用于各种ML任务,我们的方法在性能上相比强大的开源AutoML代理有4%的绝对提升,展示了其在增强代理AutoML系统中的有效性。资源可在https://github.com/jokieleung/I-MCTS上获得。

英文摘要:

Recent advancements in large language models (LLMs) have shown remarkable potential in automating machine learning tasks. However, existing LLM-based agents often struggle with low-diversity and suboptimal code generation. While recent work has introduced Monte Carlo Tree Search (MCTS) to address these issues, limitations persist in the quality and diversity of thoughts generated, as well as in the scalar value feedback mechanisms used for node selection. In this study, we introduce Introspective Monte Carlo Tree Search (I-MCTS), a novel approach that iteratively expands tree nodes through an introspective process that meticulously analyzes solutions and results from parent and sibling nodes. This facilitates a continuous refinement of the node in the search tree, thereby enhancing the overall decision-making process. Furthermore, we integrate a Large Language Model (LLM)-based value model to facilitate direct evaluation of each node's solution prior to conducting comprehensive computational rollouts. A hybrid rewarding mechanism is implemented to seamlessly transition the Q-value from LLM-estimated scores to actual performance scores. This allows higher-quality nodes to be traversed earlier. Applied to the various ML tasks, our approach demonstrates a 4% absolute improvement in performance compared to the strong open-source AutoML agents, showcasing its effectiveness in enhancing agentic AutoML systems. Resource available at https://github.com/jokieleung/I-MCTS

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。

↑