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 2302.12341stat.MLcs.LGstat.ME
收藏

基于秩的后非线性模型因果发现

Rank-Based Causal Discovery for Post-Nonlinear Models

Technical University of Munich(慕尼黑工业大学);Munich Center for Machine Learning(慕尼黑机器学习中心)

Grigor Keropyan, David Strieder, Mathias Drton

AI总结:

针对双变量情形外后非线性模型因果发现易过拟合难调参的问题,提出基于秩的方法估计函数参数,分离函数估计与独立性检验,证明一致性并经实验验证。

摘要:

从经验观测中学习因果关系是科学研究的一项核心任务。一种常见方法是采用结构因果模型,该模型假设在一组相互作用的变量之间存在带噪声的函数关系。为了确保因果方向的唯一可识别性,研究人员考虑了结构因果模型的受限子类。后非线性(PNL)因果模型是此类受限子类中最灵活的选择之一,其中特别包含了流行的加性噪声模型作为其进一步子类。然而,超越双变量情形的PNL模型学习尚未得到充分研究。现有方法通过最小化残差依赖性来学习非线性函数关系,随后对残差进行独立性检验以确定因果方向。但这些方法容易产生过拟合,因此在实践中难以适当调参。作为替代方案,我们提出了一种新的PNL因果发现方法,该方法使用基于秩的方法来估计函数参数。这种新方法利用了PNL模型的自然不变性,并将非线性函数的估计与用于寻找因果方向的独立性检验分离开来。我们证明了该方法的一致性,并在数值实验中验证了结果。

英文摘要:

Learning causal relationships from empirical observations is a central task in scientific research. A common method is to employ structural causal models that postulate noisy functional relations among a set of interacting variables. To ensure unique identifiability of causal directions, researchers consider restricted subclasses of structural causal models. Post-nonlinear (PNL) causal models constitute one of the most flexible options for such restricted subclasses, containing in particular the popular additive noise models as a further subclass. However, learning PNL models is not well studied beyond the bivariate case. The existing methods learn non-linear functional relations by minimizing residual dependencies and subsequently test independence from residuals to determine causal orientations. However, these methods can be prone to overfitting and, thus, difficult to tune appropriately in practice. As an alternative, we propose a new approach for PNL causal discovery that uses rank-based methods to estimate the functional parameters. This new approach exploits natural invariances of PNL models and disentangles the estimation of the non-linear functions from the independence tests used to find causal orientations. We prove consistency of our method and validate our results in numerical experiments.

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。

↑