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 2402.17879cs.LGcs.CL
收藏

利用语言模型实现自动化统计模型发现

Automated Statistical Model Discovery with Language Models

Stanford University(斯坦福大学);Chan Zuckerberg Biohub(陈·扎克伯格生物中心)

Michael Y. Li, Emily B. Fox, Noah D. Goodman

AI总结:

本研究提出基于大语言模型的自动化统计模型发现方法,将其置于Box's Loop框架内实现模型提出与评判的迭代,无需手工设计搜索流程,在三类概率建模场景中表现媲美专家模型且可解释拓展经典模型。

摘要:

统计模型发现是在受领域特定约束的庞大模型空间中进行的一项具有挑战性的搜索任务。要在该空间中高效搜索,需要具备建模与问题领域的专业知识。受大语言模型(LM)的领域知识与编程能力启发,我们提出了一种由语言模型驱动的自动化统计模型发现方法。我们将该自动化流程置于Box's Loop(Box循环)这一原则性框架内:LM在两种角色间迭代切换,一是作为建模者提出以概率程序表示的统计模型,二是作为领域专家对这些模型进行评判。借助LM,我们无需定义领域特定的模型语言,也无需设计手工打造的搜索流程,而这两点正是以往系统的核心局限。我们在概率建模的三种场景下评估了该方法:在受限模型空间内搜索、在开放式空间中搜索,以及在自然语言约束(例如“该模型应能被生态学家理解”)下改进专家模型。我们的方法能识别出与人类专家设计的模型性能相当的模型,还能以可解释的方式拓展经典模型。研究结果凸显了LM驱动的模型发现的应用前景。

英文摘要:

Statistical model discovery is a challenging search over a vast space of models subject to domain-specific constraints. Efficiently searching over this space requires expertise in modeling and the problem domain. Motivated by the domain knowledge and programming capabilities of large language models (LMs), we introduce a method for language model driven automated statistical model discovery. We cast our automated procedure within the principled framework of Box's Loop: the LM iterates between proposing statistical models represented as probabilistic programs, acting as a modeler, and critiquing those models, acting as a domain expert. By leveraging LMs, we do not have to define a domain-specific language of models or design a handcrafted search procedure, which are key restrictions of previous systems. We evaluate our method in three settings in probabilistic modeling: searching within a restricted space of models, searching over an open-ended space, and improving expert models under natural language constraints (e.g., this model should be interpretable to an ecologist). Our method identifies models on par with human expert designed models and extends classic models in interpretable ways. Our results highlight the promise of LM-driven model discovery.

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。

↑