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 2403.06499stat.MLcs.ITcs.LGmath.IT
收藏

基于 NML 编码的离散、混合及连续变量中未观测共同原因检测

Detection of Unobserved Common Causes based on NML Code in Discrete, Mixed, and Continuous Variables

The University of Tokyo(东京大学);IBM Research – Tokyo(IBM东京研究实验室);Graduate School of Information Science and Technology, The University of Tokyo(东京大学情报理工学研究院)

Masatoshi Kobayashi, Kohei Miyagichi, Shin Matsushima

AI总结:

本文将无需方程形式假设的因果发现方法 CLOUD 从离散数据扩展到离散、混合与连续变量,并从理论一致性及合成和真实数据实验上验证其优势。

摘要:

仅从观测数据中在存在未观测共同原因的情况下进行因果发现,是一个关键但富有挑战性的问题。我们将两个随机变量之间所有可能的因果关系划分为以下四类,并旨在从观测数据中识别其中一类:两种存在任一方向直接因果关系的情形、变量相互独立的情形,以及变量受到潜在混杂因子混杂的情形。尽管已有方法被提出以解决该问题,但它们要求未观测变量满足关于其方程模型形式的假设。在我们此前的研究(Kobayashi 等人,2022)中,首次提出了一种无需此类假设、适用于离散数据的因果发现方法,并将其命名为 CLOUD。利用归一化最大似然(Normalized Maximum Likelihood,NML)编码,CLOUD 从一组候选模型中选择使观测数据码长最小的模型。本文将 CLOUD 扩展,使其适用于离散、混合和连续等多种数据类型。我们不仅通过理论分析证明了 CLOUD 在模型选择方面的一致性,还通过在合成数据和真实世界数据上的大量实验,证明 CLOUD 在推断因果关系方面比现有方法更有效。

英文摘要:

Causal discovery in the presence of unobserved common causes from observational data only is a crucial but challenging problem. We categorize all possible causal relationships between two random variables into the following four categories and aim to identify one from observed data: two cases in which either of the direct causality exists, a case that variables are independent, and a case that variables are confounded by latent confounders. Although existing methods have been proposed to tackle this problem, they require unobserved variables to satisfy assumptions on the form of their equation models. In our previous study (Kobayashi et al., 2022), the first causal discovery method without such assumptions is proposed for discrete data and named CLOUD. Using Normalized Maximum Likelihood (NML) Code, CLOUD selects a model that yields the minimum codelength of the observed data from a set of model candidates. This paper extends CLOUD to apply for various data types across discrete, mixed, and continuous. We not only performed theoretical analysis to show the consistency of CLOUD in terms of the model selection, but also demonstrated that CLOUD is more effective than existing methods in inferring causal relationships by extensive experiments on both synthetic and real-world data.

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。

↑