TY - GEN
T1 - Causal mediation analysis with hidden confounders
AU - Cheng, Lu
AU - Guo, Ruocheng
AU - Liu, Huan
N1 - Funding Information:
This work is supported by ONR N00014-21-1-4002 and ARO W911NF 2110030. The views, opinions and/or findings expressed are the authors’ and should not be interpreted as representing the official views or policies of the ARO or ONR.
Publisher Copyright:
© 2022 ACM.
PY - 2022/2/11
Y1 - 2022/2/11
N2 - An important problem in causal inference is to break down the total effect of a treatment on an outcome into different causal pathways and to quantify the causal effect in each pathway. For instance, in causal fairness, the total effect of being a male employee (i.e., treatment) constitutes its direct effect on annual income (i.e., outcome) and the indirect effect via the employee's occupation (i.e., mediator). Causal mediation analysis (CMA) is a formal statistical framework commonly used to reveal such underlying causal mechanisms. One major challenge of CMA in observational studies is handling confounders, variables that cause spurious causal relationships among treatment, mediator, and outcome. Conventional methods assume sequential ignorability that implies all confounders can be measured, which is often unverifiable in practice. This work aims to circumvent the stringent sequential ignorability assumptions and consider hidden confounders. Drawing upon proxy strategies and recent advances in deep learning, we propose to simultaneously uncover the latent variables that characterize hidden confounders and estimate the causal effects. Empirical evaluations using both synthetic and semi-synthetic datasets validate the effectiveness of the proposed method. We further show the potentials of our approach for causal fairness analysis.
AB - An important problem in causal inference is to break down the total effect of a treatment on an outcome into different causal pathways and to quantify the causal effect in each pathway. For instance, in causal fairness, the total effect of being a male employee (i.e., treatment) constitutes its direct effect on annual income (i.e., outcome) and the indirect effect via the employee's occupation (i.e., mediator). Causal mediation analysis (CMA) is a formal statistical framework commonly used to reveal such underlying causal mechanisms. One major challenge of CMA in observational studies is handling confounders, variables that cause spurious causal relationships among treatment, mediator, and outcome. Conventional methods assume sequential ignorability that implies all confounders can be measured, which is often unverifiable in practice. This work aims to circumvent the stringent sequential ignorability assumptions and consider hidden confounders. Drawing upon proxy strategies and recent advances in deep learning, we propose to simultaneously uncover the latent variables that characterize hidden confounders and estimate the causal effects. Empirical evaluations using both synthetic and semi-synthetic datasets validate the effectiveness of the proposed method. We further show the potentials of our approach for causal fairness analysis.
KW - Causal mediation analysis
KW - Confounders
KW - Fairness
KW - Latent-variable model
KW - Proxy variable
UR - http://www.scopus.com/inward/record.url?scp=85125786714&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85125786714&partnerID=8YFLogxK
U2 - 10.1145/3488560.3498407
DO - 10.1145/3488560.3498407
M3 - Conference contribution
AN - SCOPUS:85125786714
T3 - WSDM 2022 - Proceedings of the 15th ACM International Conference on Web Search and Data Mining
SP - 113
EP - 122
BT - WSDM 2022 - Proceedings of the 15th ACM International Conference on Web Search and Data Mining
PB - Association for Computing Machinery, Inc
T2 - 15th ACM International Conference on Web Search and Data Mining, WSDM 2022
Y2 - 21 February 2022 through 25 February 2022
ER -