A0255
Title: Conditional generative adversarial networks for individualized causal mediation analysis
Authors: Xinyuan Song - Chinese University of Hong Kong (Hong Kong) [presenting]
Abstract: Most classical methods popular in causal mediation analysis can only estimate the average causal effects and are difficult to apply to precision medicine. Although identifying heterogeneous causal effects has received some attention, the causal effects are explored using assumptive parametric models with limited model flexibility and analytic power. Recently, machine learning has become a major tool for accurately estimating individualized causal effects, thanks to its flexibility in model forms and efficiency in capturing complex nonlinear relationships. A novel method, conditional generative adversarial network (CGAN), is proposed for individualized causal mediation analysis (CGAN-ICMA) to infer individualized causal effects based on the CGAN framework. Simulation studies show that CGANICMA outperforms five other state-of-the-art methods, including linear regression, k-nearest neighbor, support vector machine regression, decision tree, and random forest regression. The proposed model is then applied to a study on the Alzheimer's Disease Neuroimaging Initiative dataset. The application further demonstrates the utility of the proposed method in estimating the individualized causal effects of the APOE-"4 alleles on cognitive impairment directly or through mediators.
