Proximal Learning for Individualized Treatment Regimes Under Unmeasured Confounding

Data-driven individualized decision making has recently received increasing research interests. Most existing methods rely on the assumption of no unmeasured confounding, which unfortunately cannot be ensured in practice especially in observational studies. Motivated by the recent proposed proximal causal inference, we develop several proximal learning approaches to estimating optimal individualized treatment regimes (ITRs) in the presence of unmeasured confounding. In particular, we establish several identification results for different classes of ITRs, exhibiting the trade-off between the risk of making untestable assumptions and the value function improvement in decision making. Based on these results, we propose several classification-based approaches to finding a variety of restricted in-class optimal ITRs and develop their theoretical properties. The appealing numerical performance of our proposed methods is demonstrated via an extensive simulation study and one real data application. keywords: Individualized treatment regimes, Unmeasured confounding, Proximal causal inference, Regime identification Department of Decision Sciences, The George Washington University Department of Statistics, The George Washington University 1

[1]  R. Kronmal,et al.  Assessing the sensitivity of regression results to unmeasured confounders in observational studies. , 1998, Biometrics.

[2]  W. Miao,et al.  A Confounding Bridge Approach for Double Negative Control Inference on Causal Effects (Supplement and Sample Codes are included) , 2018, 1808.04945.

[3]  Eric B. Laber,et al.  A Robust Method for Estimating Optimal Treatment Regimes , 2012, Biometrics.

[4]  Michael R Kosorok,et al.  Residual Weighted Learning for Estimating Individualized Treatment Rules , 2015, Journal of the American Statistical Association.

[5]  Donglin Zeng,et al.  Estimating Individualized Treatment Rules Using Outcome Weighted Learning , 2012, Journal of the American Statistical Association.

[6]  C. O'connor,et al.  Standardized Team-Based Care for Cardiogenic Shock. , 2019, Journal of the American College of Cardiology.

[7]  Michael R. Kosorok,et al.  High-Dimensional Precision Medicine From Patient-Derived Xenografts , 2019, Journal of the American Statistical Association.

[8]  Peter Dayan,et al.  Q-learning , 1992, Machine Learning.

[9]  W. Newey,et al.  Local Identification of Nonparametric and Semiparametric Models , 2011, 1105.3007.

[10]  Ilya Shpitser,et al.  Minimax Kernel Machine Learning for a Class of Doubly Robust Functionals , 2021, ArXiv.

[11]  Stefan Wager,et al.  Policy Learning With Observational Data , 2017, Econometrica.

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

[13]  Masatoshi Uehara,et al.  Causal Inference Under Unmeasured Confounding With Negative Controls: A Minimax Learning Approach , 2021, ArXiv.

[14]  L. Stevenson,et al.  Trends in Utilization and Outcomes of Pulmonary Artery Catheterization in Heart Failure With and Without Cardiogenic Shock. , 2019, Journal of cardiac failure.

[15]  I. König,et al.  What is precision medicine? , 2017, European Respiratory Journal.

[16]  Rong Jin,et al.  Nyström Method vs Random Fourier Features: A Theoretical and Empirical Comparison , 2012, NIPS.

[17]  Xiaohong Chen,et al.  Semi‐Nonparametric IV Estimation of Shape‐Invariant Engel Curves , 2003 .

[18]  E. J. Tchetgen Tchetgen,et al.  An Introduction to Proximal Causal Learning , 2020, medRxiv.

[19]  W. Newey,et al.  Instrumental variable estimation of nonparametric models , 2003 .

[20]  Sukjin Han Optimal Dynamic Treatment Regimes and Partial Welfare Ordering , 2019, Journal of the American Statistical Association.

[21]  Eric B. Laber,et al.  Doubly Robust Learning for Estimating Individualized Treatment with Censored Data. , 2015, Biometrika.

[22]  J. Robins A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , 1986 .

[23]  Chengchun Shi,et al.  HIGH-DIMENSIONAL A-LEARNING FOR OPTIMAL DYNAMIC TREATMENT REGIMES. , 2018, Annals of statistics.

[24]  Yifan Cui,et al.  A Semiparametric Instrumental Variable Approach to Optimal Treatment Regimes Under Endogeneity , 2019, Journal of the American Statistical Association.

[25]  Stijn Vansteelandt,et al.  Bias-Reduced Doubly Robust Estimation , 2015 .

[26]  S. Murphy,et al.  PERFORMANCE GUARANTEES FOR INDIVIDUALIZED TREATMENT RULES. , 2011, Annals of statistics.

[27]  John Langford,et al.  Doubly Robust Policy Evaluation and Learning , 2011, ICML.

[28]  Z. Geng,et al.  Identifying Causal Effects With Proxy Variables of an Unmeasured Confounder. , 2016, Biometrika.

[29]  Le Song,et al.  Feature Selection via Dependence Maximization , 2012, J. Mach. Learn. Res..

[30]  M. Kosorok,et al.  Reinforcement learning design for cancer clinical trials , 2009, Statistics in medicine.

[31]  Eric J. Tchetgen Tchetgen,et al.  Multiply robust causal inference with double‐negative control adjustment for categorical unmeasured confounding , 2018, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[32]  Bharath Bhushan Damodaran Fast Optimal Bandwidth Selection for RBF Kernel using Reproducing Kernel Hilbert Space Operators for Kernel Based Classifiers , 2018, ArXiv.

[33]  Raymond K. W. Wong,et al.  Kernel-based covariate functional balancing for observational studies. , 2018, Biometrika.

[34]  Zhiqiang Tan,et al.  A Distributional Approach for Causal Inference Using Propensity Scores , 2006 .

[35]  John M. Cioffi,et al.  Optimal Resource Allocation for OFDMA Downlink Systems , 2006, 2006 IEEE International Symposium on Information Theory.

[36]  Gary Chamberlain,et al.  Efficiency Bounds for Semiparametric Regression , 1992 .

[37]  Ambuj Tewari,et al.  Microrandomized trials: An experimental design for developing just-in-time adaptive interventions. , 2015, Health psychology : official journal of the Division of Health Psychology, American Psychological Association.

[38]  Xiaohong Chen,et al.  Efficient Estimation of Models with Conditional Moment Restrictions Containing Unknown Functions , 2003 .

[39]  Nishanth Dikkala,et al.  Minimax Estimation of Conditional Moment Models , 2020, NeurIPS.

[40]  Donglin Zeng,et al.  Personalized Dose Finding Using Outcome Weighted Learning , 2016, Journal of the American Statistical Association.

[41]  Rui Song,et al.  Quantile-Optimal Treatment Regimes , 2018, Journal of the American Statistical Association.

[42]  L. Goldman,et al.  The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. , 1996, JAMA.

[43]  Bo Zhang,et al.  Estimating Optimal Treatment Rules with an Instrumental Variable: A Semi-Supervised Learning Approach , 2020 .

[44]  Yang Ning,et al.  Efficient augmentation and relaxation learning for individualized treatment rules using observational data , 2019, J. Mach. Learn. Res..

[45]  Kengo Kato,et al.  Gaussian approximation of suprema of empirical processes , 2012, 1212.6885.

[46]  Yifan Cui,et al.  Semiparametric proximal causal inference , 2020 .

[47]  J. Robins,et al.  Double/Debiased Machine Learning for Treatment and Structural Parameters , 2017 .

[48]  Xiaohong Chen,et al.  Estimation of Nonparametric Conditional Moment Models with Possibly Nonsmooth Generalized Residuals , 2009 .

[49]  R. Hogg,et al.  On adaptive estimation , 1984 .

[50]  Jan Peters,et al.  Reinforcement learning in robotics: A survey , 2013, Int. J. Robotics Res..

[51]  Xavier D'Haultfoeuille,et al.  ON THE COMPLETENESS CONDITION IN NONPARAMETRIC INSTRUMENTAL PROBLEMS , 2010, Econometric Theory.

[52]  C. Manski Statistical treatment rules for heterogeneous populations , 2003 .

[53]  Bernhard Schölkopf,et al.  Kernel Choice and Classifiability for RKHS Embeddings of Probability Distributions , 2009, NIPS.

[54]  Marco Carone,et al.  Optimal Individualized Decision Rules Using Instrumental Variable Methods , 2020, Journal of the American Statistical Association.

[55]  A. Ciarka,et al.  Right Heart Catheterization—Background, Physiological Basics, and Clinical Implications , 2019, Journal of clinical medicine.