Hi-CI: Deep Causal Inference in High Dimensions

We address the problem of counterfactual regression using causal inference (CI) in observational studies consisting of high dimensional covariates and high cardinality treatments. Confounding bias, which leads to inaccurate treatment effect estimation, is attributed to covariates that affect both treatments and outcome. The presence of high-dimensional co-variates exacerbates the impact of bias as it is harder to isolate and measure the impact of these confounders. In the presence of high-cardinality treatment variables, CI is rendered ill-posed due to the increase in the number of counterfactual outcomes to be predicted. We propose Hi-CI, a deep neural network (DNN) based framework for estimating causal effects in the presence of large number of covariates, and high-cardinal and continuous treatment variables. The proposed architecture comprises of a decorrelation network and an outcome prediction network. In the decorrelation network, we learn a data representation in lower dimensions as compared to the original covariates and addresses confounding bias alongside. Subsequently, in the outcome prediction network, we learn an embedding of high-cardinality and continuous treatments, jointly with the data representation. We demonstrate the efficacy of causal effect prediction of the proposed Hi-CI network using synthetic and real-world NEWS datasets.

[1]  Uri Shalit,et al.  Estimating individual treatment effect: generalization bounds and algorithms , 2016, ICML.

[2]  J. Robins,et al.  Estimation of Regression Coefficients When Some Regressors are not Always Observed , 1994 .

[3]  Peter Bühlmann,et al.  Causal statistical inference in high dimensions , 2013, Math. Methods Oper. Res..

[4]  Lovekesh Vig,et al.  MetaCI: Meta-Learning for Causal Inference in a Heterogeneous Population , 2019, ArXiv.

[5]  Lovekesh Vig,et al.  MultiMBNN: Matched and Balanced Causal Inference with Neural Networks , 2020, ESANN.

[6]  Isaac Dialsingh,et al.  Applied Bayesian Modeling and Causal Inference from Incomplete Data Perspectives , 2005 .

[7]  Joaquin Quiñonero Candela,et al.  Counterfactual reasoning and learning systems: the example of computational advertising , 2013, J. Mach. Learn. Res..

[8]  Chad Hazlett,et al.  Covariate balancing propensity score for a continuous treatment: Application to the efficacy of political advertisements , 2018 .

[9]  Zichen Zhang,et al.  Reducing Selection Bias in Counterfactual Reasoning for Individual Treatment Effects Estimation , 2019, ArXiv.

[10]  Foster Provost,et al.  Causally motivated attribution for online advertising , 2012, ADKDD '12.

[11]  Victor Chernozhukov,et al.  Inference on Treatment Effects after Selection Amongst High-Dimensional Controls , 2011 .

[12]  L. Xue,et al.  CBPS-Based Inference in Nonlinear Regression Models with Missing Data , 2016 .

[13]  Mihaela van der Schaar,et al.  Deep-Treat: Learning Optimal Personalized Treatments From Observational Data Using Neural Networks , 2018, AAAI.

[14]  Eustache Diemert,et al.  Attribution Modeling Increases Efficiency of Bidding in Display Advertising , 2017, ADKDD@KDD.

[15]  Jianqing Fan,et al.  Improving Covariate Balancing Propensity Score : A Doubly Robust and Efficient Approach ∗ , 2016 .

[16]  Uri Shalit,et al.  Learning Representations for Counterfactual Inference , 2016, ICML.

[17]  Gary King,et al.  Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference , 2007, Political Analysis.

[18]  Vikas Ramachandra,et al.  Deep Learning for Causal Inference , 2018, ArXiv.

[19]  G. Imbens,et al.  The Propensity Score with Continuous Treatments , 2005 .

[20]  K. Imai,et al.  Covariate balancing propensity score , 2014 .

[21]  Walter Karlen,et al.  Perfect Match: A Simple Method for Learning Representations For Counterfactual Inference With Neural Networks , 2018, ArXiv.

[22]  Stefan Bauer,et al.  Learning Counterfactual Representations for Estimating Individual Dose-Response Curves , 2019, AAAI.

[23]  A. Belloni,et al.  Inference on Treatment Effects after Selection Amongst High-Dimensional Controls , 2011, 1201.0224.

[24]  Douglas Galagate,et al.  Causal inference with a continuous treatment and outcome: Alternative estimators for parametric dose-response functions with applications , 2016 .

[25]  G. Imbens,et al.  Approximate residual balancing: debiased inference of average treatment effects in high dimensions , 2016, 1604.07125.

[26]  Jian Yang,et al.  Causal Inference via Sparse Additive Models with Application to Online Advertising , 2015, AAAI.

[27]  Joshua M. Stuart,et al.  The Cancer Genome Atlas Pan-Cancer analysis project , 2013, Nature Genetics.

[28]  Edward H Kennedy,et al.  Non‐parametric methods for doubly robust estimation of continuous treatment effects , 2015, Journal of the Royal Statistical Society. Series B, Statistical methodology.