Estimating Causal Effects on Networked Observational Data via Representation Learning

被引:12
作者
Jiang, Song [1 ]
Sun, Yizhou [1 ]
机构
[1] Univ Calif Los Angeles, Los Angeles, CA USA
来源
PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2022 | 2022年
关键词
causal inference; network effect; graph neural networks; INFERENCE; DIAGRAMS;
D O I
10.1145/3511808.3557311
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we study the causal effects estimation problem on networked observational data. We theoretically prove that standard graph machine learning (ML) models, e.g., graph neural networks (GNNs), fail in estimating the causal effects on networks. We show that graph ML models exhibit two distribution mismatches of their objective functions compared to causal effects estimation, leading to the failure of traditional ML models. Motivated by this, we first formulate the networked causal effects estimation as a data-driven multi-task learning problem, and then propose a novel framework NetEst to conduct causal inference in the network setting. NetEst uses GNNs to learn representations for confounders, which are from both a unit's own characteristics and the network effects. The embeddings are then used to sufficiently bridge the distribution gaps via adversarial learning and estimate the observed outcomes simultaneously. Extensive experimental studies on two real-world networks with semi-synthetic data demonstrate the effectiveness of NetEst. We also provide analyses on why and when NetEst works.
引用
收藏
页码:852 / 861
页数:10
相关论文
共 45 条
[1]   VACCINATION AND HERD-IMMUNITY TO INFECTIOUS-DISEASES [J].
ANDERSON, RM ;
MAY, RM .
NATURE, 1985, 318 (6044) :323-329
[2]   Inferring Network Effects from Observational Data [J].
Arbour, David ;
Garant, Dan ;
Jensen, David .
KDD'16: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2016, :715-724
[3]   Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies [J].
Austin, Peter C. ;
Stuart, Elizabeth A. .
STATISTICS IN MEDICINE, 2015, 34 (28) :3661-3679
[4]   CAUSAL INFERENCE FROM OBSERVATIONAL STUDIES WITH CLUSTERED INTERFERENCE, WITH APPLICATION TO A CHOLERA VACCINE STUDY [J].
Barkley, Brian G. ;
Hudgens, Michael G. ;
Clemens, John D. ;
Ali, Mohammad ;
Emch, Michael E. .
ANNALS OF APPLIED STATISTICS, 2020, 14 (03) :1432-1448
[5]   Latent Dirichlet allocation [J].
Blei, DM ;
Ng, AY ;
Jordan, MI .
JOURNAL OF MACHINE LEARNING RESEARCH, 2003, 3 (4-5) :993-1022
[6]   Graph Infomax Adversarial Learning for Treatment Effect Estimation with Networked Observational Data [J].
Chu, Zhixuan ;
Rathbun, Stephen L. ;
Li, Sheng .
KDD '21: PROCEEDINGS OF THE 27TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2021, :176-184
[7]  
Johansson FD, 2018, Arxiv, DOI arXiv:1802.08598
[8]  
Diederik P., 2015, INT C LEARN REPR ICL
[9]   Identification and Estimation of Treatment and Interference Effects in Observational Studies on Networks [J].
Forastiere, Laura ;
Airoldi, Edoardo M. ;
Mealli, Fabrizia .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2021, 116 (534) :901-918
[10]  
Goodfellow I. J., 2014, arXiv, DOI DOI 10.48550/ARXIV.1406.2661