[论文解读] Causal Discovery from a Mixture of Experimental and Observational Data

Gregory F. Cooper, Changwon Yoo|arXiv (Cornell University)|Jan 23, 2013

Bayesian Modeling and Causal Inference参考文献 7被引用 172

一句话总结

本文提出了一种贝叶斯方法，用于从混合的实验数据和观测数据中进行因果发现，通过整合两种数据类型来学习因果贝叶斯网络。通过利用干预（实验数据）和被动观测，该方法在重构因果结构和估计参数方面提高了准确性，如在ALARM网络上通过系统性地改变数据比例所展示的，结构恢复和参数估计均取得了显著提升。

ABSTRACT

This paper describes a Bayesian method for combining an arbitrary mixture of observational and experimental data in order to learn causal Bayesian networks. Observational data are passively observed. Experimental data, such as that produced by randomized controlled trials, result from the experimenter manipulating one or more variables (typically randomly) and observing the states of other variables. The paper presents a Bayesian method for learning the causal structure and parameters of the underlying causal process that is generating the data, given that (1) the data contains a mixture of observational and experimental case records, and (2) the causal process is modeled as a causal Bayesian network. This learning method was applied using as input various mixtures of experimental and observational data that were generated from the ALARM causal Bayesian network. In these experiments, the absolute and relative quantities of experimental and observational data were varied systematically. For each of these training datasets, the learning method was applied to predict the causal structure and to estimate the causal parameters that exist among randomly selected pairs of nodes in ALARM that are not confounded. The paper reports how these structure predictions and parameter estimates compare with the true causal structures and parameters as given by the ALARM network.

研究动机与目标

• 开发一种结合实验数据和观测数据的方法，以在贝叶斯网络中实现更优的因果发现。
• 解决在数据来源异质（包括基于干预和被动观测记录）的情况下学习因果结构的挑战。
• 评估实验数据与观测数据比例变化对因果结构和参数学习准确性的影响。
• 在受控数据混合条件下，于基准因果网络（ALARM）上展示该方法的有效性。

提出的方法

• 该方法采用贝叶斯框架，联合从观测数据和实验数据的混合中学习因果结构和参数。
• 其将数据生成过程建模为因果贝叶斯网络，同时纳入被动观测和基于干预的数据。
• 该方法使用条件概率分布来表示因果关系，并将干预显式建模为do-操作。
• 学习算法使用贝叶斯定理，计算可能因果结构和参数的后验分布，整合两种数据类型提供的证据。
• 通过在评估期间聚焦于ALARM网络中非混淆的节点对，该方法处理了混杂因素的影响。
• 通过系统性地改变实验数据与观测数据的相对数量，评估不同数据混合下的性能表现。

实验结果

研究问题

• RQ1 与单独使用任一数据类型相比，结合实验数据和观测数据在因果结构发现方面有何改进？
• RQ2 实验数据比例的变化对因果参数估计准确性有何影响？
• RQ3 当提供混合数据时，该方法能否可靠地恢复已知网络（ALARM）的真实因果结构？
• RQ4 干预的存在如何影响贝叶斯学习过程中正确因果结构的后验概率？

主要发现

• 即使实验数据比例较小，包含实验数据也能显著提升因果结构的恢复效果。
• 在引入实验数据后，参数估计更加准确，尤其在直接因果效应的估计上表现更优。
• 该方法在ALARM网络中对非混淆节点对的真正因果关系识别达到了高准确度。
• 随着实验数据比例的增加，性能持续提升，证明了干预的价值。