A Unified Survey of Treatment Effect Heterogeneity Modelling and Uplift Modelling

A central question in many fields of scientific research is to determine how an outcome is affected by an action, i.e., to estimate the causal effect or treatment effect of an action. In recent years, in areas such as personalised healthcare, sociology, and online marketing, a need has emerged to estimate heterogeneous treatment effects with respect to individuals of different characteristics. To meet this need, two major approaches have been taken: treatment effect heterogeneity modelling and uplifting modelling. Researchers and practitioners in different communities have developed algorithms based on these approaches to estimate the heterogeneous treatment effects. In this article, we present a unified view of these two seemingly disconnected yet closely related approaches under the potential outcome framework. We provide a structured survey of existing methods following either of the two approaches, emphasising their inherent connections and using unified notation to facilitate comparisons. We also review the main applications of the surveyed methods in personalised marketing, personalised medicine, and sociology. Finally, we summarise and discuss the available software packages and source codes in terms of their coverage of different methods and applicability to different datasets, and we provide general guidelines for method selection.

[1]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[2]  Michael Lechner,et al.  Machine Learning Estimation of Heterogeneous Causal Effects: Empirical Monte Carlo Evidence , 2018, The Econometrics Journal.

[3]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[4]  Jennifer Hill,et al.  Automated versus Do-It-Yourself Methods for Causal Inference: Lessons Learned from a Data Analysis Competition , 2017, Statistical Science.

[5]  J. Bellon.  Personalized Radiation Oncology for Breast Cancer: The New Frontier. , 2015, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[6]  Denis Larocque,et al.  Non-parametric individual treatment effect estimation for survival data with random forests , 2019, Bioinform..

[7]  K. Imai,et al.  Propensity-Score Based Methods for Causal Inference in Observational Studies with Fixed Non-Binary Treatments Shandong , 2013 .

[8]  T. Richardson,et al.  Covariate selection for the nonparametric estimation of an average treatment effect , 2011 .

[9]  G. Imbens,et al.  Machine Learning for Estimating Heterogeneous Causal Effects , 2015 .

[10]  R. Lalonde.  Evaluating the Econometric Evaluations of Training Programs with Experimental Data , 1984 .

[11]  Behram Hansotia,et al.  Incremental value modeling , 2002 .

[12]  Wouter Verbeke,et al.  A survey and benchmarking study of multitreatment uplift modeling , 2020, Data Mining and Knowledge Discovery.

[13]  Leo Guelman,et al.  Optimal personalized treatment rules for marketing interventions: A review of methods, a new proposal, and an insurance case study , 2014 .

[14]  D. Schoenfeld.  The asymptotic properties of nonparametric tests for comparing survival distributions , 1981 .

[15]  Jenny Häggström,et al.  Data‐driven confounder selection via Markov and Bayesian networks , 2016, Biometrics.

[16]  Christopher Winship,et al.  Counterfactuals and Causal Inference: Methods and Principles for Social Research , 2007 .

[17]  Szymon Jaroszewicz,et al.  Decision trees for uplift modeling with single and multiple treatments , 2011, Knowledge and Information Systems.

[18]  Wouter Verbeke,et al.  A Literature Survey and Experimental Evaluation of the State-of-the-Art in Uplift Modeling: A Stepping Stone Toward the Development of Prescriptive Analytics , 2018, Big Data.

[19]  Susan Athey,et al.  Beyond prediction: Using big data for policy problems , 2017, Science.

[20]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[21]  Vladimir Vapnik,et al.  Support-vector networks , 2004, Machine Learning.

[22]  Lin Liu,et al.  Estimating heterogeneous treatment effect by balancing heterogeneity and fitness , 2018, BMC Bioinformatics.

[23]  Donald Rubin,et al.  Estimating Causal Effects from Large Data Sets Using Propensity Scores , 1997, Annals of Internal Medicine.

[24]  David Maxwell Chickering,et al.  A Decision Theoretic Approach to Targeted Advertising , 2000, UAI.

[25]  Xiaojing Li,et al.  A joint optimization of incrementality and revenue to satisfy both advertiser and publisher , 2013, WWW.

[26]  Avi Feller,et al.  Assessing Treatment Effect Variation in Observational Studies: Results from a Data Challenge , 2019, Observational Studies.

[27]  Thorsten Joachims,et al.  A support vector method for multivariate performance measures , 2005, ICML.

[28]  R. Tibshirani,et al.  Semi-Supervised Methods to Predict Patient Survival from Gene Expression Data , 2004, PLoS biology.

[29]  Jinsung Yoon,et al.  Discovery and Clinical Decision Support for Personalized Healthcare , 2017, IEEE Journal of Biomedical and Health Informatics.

[30]  Leo Guelman,et al.  A decision support framework to implement optimal personalized marketing interventions , 2015, Decis. Support Syst..

[31]  L. Tian,et al.  Analysis of randomized comparative clinical trial data for personalized treatment selections. , 2011, Biostatistics.

[32]  Nicholas Radcliffe,et al.  Using control groups to target on predicted lift: Building and assessing uplift model , 2007 .

[33]  Xin Yan,et al.  Facilitating score and causal inference trees for large observational studies , 2012, J. Mach. Learn. Res..

[34]  M A Fischl,et al.  A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. AIDS Clinical Trials Group 320 Study Team. , 1997, The New England journal of medicine.

[35]  Tobias von Martens,et al.  Effective customer selection for marketing campaigns based on net scores , 2017 .

[36]  Szymon Jaroszewicz,et al.  Ensemble methods for uplift modeling , 2014, Data Mining and Knowledge Discovery.

[37]  Bernhard Schölkopf,et al.  Elements of Causal Inference: Foundations and Learning Algorithms , 2017 .

[38]  D. Rubin,et al.  Causal Inference for Statistics, Social, and Biomedical Sciences: Sensitivity Analysis and Bounds , 2015 .

[39]  David E. Broockman,et al.  Durably reducing transphobia: A field experiment on door-to-door canvassing , 2016, Science.

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

[41]  Negar Hassanpour,et al.  CounterFactual Regression with Importance Sampling Weights , 2019, IJCAI.

[42]  Xiaogang Su,et al.  Subgroup Analysis via Recursive Partitioning , 2009 .

[43]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[44]  Judea Pearl.  Causality by Judea Pearl , 2009 .

[45]  Alberto Maria Segre,et al.  Programs for Machine Learning , 1994 .

[46]  D. McCaffrey,et al.  Propensity score estimation with boosted regression for evaluating causal effects in observational studies. , 2004, Psychological methods.

[47]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[48]  Wray L. Buntine,et al.  Learning classification trees , 1992 .

[49]  Patrick D. Surry,et al.  Real-World Uplift Modelling with Significance-Based Uplift Trees , 2012 .

[50]  David Maxwell Chickering,et al.  Large-Sample Learning of Bayesian Networks is NP-Hard , 2002, J. Mach. Learn. Res..

[51]  Ruocheng Guo,et al.  A Survey of Learning Causality with Data , 2018, ACM Comput. Surv..

[52]  Victor S. Y. Lo.  The true lift model: a novel data mining approach to response modeling in database marketing , 2002, SKDD.

[53]  G. Imbens,et al.  Large Sample Properties of Matching Estimators for Average Treatment Effects , 2004 .

[54]  Stefan Wager,et al.  Estimation and Inference of Heterogeneous Treatment Effects using Random Forests , 2015, Journal of the American Statistical Association.

[55]  LEO GUELMAN,et al.  Uplift Random Forests , 2015, Cybern. Syst..

[56]  David Page,et al.  Score As You Lift (SAYL): A Statistical Relational Learning Approach to Uplift Modeling , 2013, ECML/PKDD.

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

[58]  Leo Guelman,et al.  Random Forests for Uplift Modeling: An Insurance Customer Retention Case , 2012, MS.

[59]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[60]  Szymon Jaroszewicz,et al.  Decision Trees for Uplift Modeling , 2010, 2010 IEEE International Conference on Data Mining.

[61]  David Page,et al.  Support Vector Machines for Differential Prediction , 2014, ECML/PKDD.

[62]  Szymon Jaroszewicz,et al.  Lp\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_p$$\end{document}-Support vector machines for uplift modeling , 2017, Knowledge and Information Systems.

[63]  Marc Ratkovic,et al.  Estimating treatment effect heterogeneity in randomized program evaluation , 2013, 1305.5682.

[64]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[65]  Sören R. Künzel,et al.  Metalearners for estimating heterogeneous treatment effects using machine learning , 2017, Proceedings of the National Academy of Sciences.

[66]  S. Athey,et al.  Estimating Treatment Effects with Causal Forests: An Application , 2019, Observational Studies.

[67]  David Page,et al.  Logical Differential Prediction Bayes Net, improving breast cancer diagnosis for older women , 2012, AMIA.

[68]  Jiuyong Li,et al.  A general framework for causal classification , 2020, International Journal of Data Science and Analytics.

[69]  Stefan Lessmann,et al.  Conversion Uplift in E-Commerce: A Systematic Benchmark of Modeling Strategies , 2019, Int. J. Inf. Technol. Decis. Mak..

[70]  I. Shpitser,et al.  A New Criterion for Confounder Selection , 2011, Biometrics.

[71]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[72]  Masashi Sugiyama,et al.  Uplift Modeling from Separate Labels , 2018, NeurIPS.

[73]  H. Weisberg,et al.  Post hoc subgroups in clinical trials: Anathema or analytics? , 2015, Clinical trials.

[74]  Susan Athey,et al.  Recursive partitioning for heterogeneous causal effects , 2015, Proceedings of the National Academy of Sciences.

[75]  Jennifer L. Hill,et al.  Bayesian Nonparametric Modeling for Causal Inference , 2011 .

[76]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[77]  Szymon Jaroszewicz,et al.  Uplift Modeling in Direct Marketing , 2012 .

[78]  B. Pitt,et al.  The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study Investigators. , 1999, The New England journal of medicine.

[79]  Lin Liu,et al.  Treatment effect estimation with disentangled latent factors , 2020, AAAI.

[80]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[81]  Patrik O. Hoyer,et al.  Data-driven covariate selection for nonparametric estimation of causal effects , 2013, AISTATS.

[82]  B. Pitt,et al.  The Effect of Spironolactone on Morbidity and Mortality in Patients with Severe Heart Failure , 2000 .

[83]  Xinkun Nie,et al.  Quasi-oracle estimation of heterogeneous treatment effects , 2017, Biometrika.

[84]  E. Diemert.  A Large Scale Benchmark for Uplift Modeling , 2018 .

[85]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[86]  S. Jaroszewicz,et al.  Uplift modeling for clinical trial data , 2012 .

[87]  Szymon Jaroszewicz,et al.  Székely Regularization for Uplift Modeling , 2016, Challenges in Computational Statistics and Data Mining.

[88]  P. Austin.  An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies , 2011, Multivariate behavioral research.

[89]  Max Welling,et al.  Causal Effect Inference with Deep Latent-Variable Models , 2017, NIPS 2017.

[90]  S. Schneeweiss,et al.  Evaluating uses of data mining techniques in propensity score estimation: a simulation study , 2008, Pharmacoepidemiology and drug safety.

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

[92]  Stan Matwin,et al.  Challenges in Computational Statistics and Data Mining , 2015, Challenges in Computational Statistics and Data Mining.

[93]  Zhi-Hua Zhou,et al.  Mining heterogeneous causal effects for personalized cancer treatment , 2017, Bioinform..

[94]  Aidong Zhang,et al.  Representation Learning for Treatment Effect Estimation from Observational Data , 2018, NeurIPS.

[95]  Kathleen Kane,et al.  Mining for the truly responsive customers and prospects using true-lift modeling: Comparison of new and existing methods , 2014 .

[96]  D. Green,et al.  The Effects of Canvassing, Telephone Calls, and Direct Mail on Voter Turnout: A Field Experiment , 2000, American Political Science Review.

[97]  J. Goeman.  L1 Penalized Estimation in the Cox Proportional Hazards Model , 2009, Biometrical journal. Biometrische Zeitschrift.

[98]  Jascha Sohl-Dickstein,et al.  Sensitivity and Generalization in Neural Networks: an Empirical Study , 2018, ICLR.

[99]  Mihaela van der Schaar,et al.  GANITE: Estimation of Individualized Treatment Effects using Generative Adversarial Nets , 2018, ICLR.

[100]  J. Brooks-Gunn,et al.  Effects of Early Intervention on Cognitive Function of Low Birth Weight Preterm Infants, , 1992, The Journal of pediatrics.

[101]  Hansheng Wang,et al.  Subgroup Analysis via Recursive Partitioning , 2009, J. Mach. Learn. Res..

[102]  VincentPascal,et al.  Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion , 2010 .

[103]  David Page,et al.  Relational Differential Prediction , 2012, ECML/PKDD.

[104]  Szymon Jaroszewicz,et al.  Support Vector Machines for Uplift Modeling , 2013, 2013 IEEE 13th International Conference on Data Mining Workshops.

[105]  Elizabeth Burnside,et al.  Learning Bayesian networks of rules with SAYU , 2005, MRDM '05.

[106]  Pierre Gutierrez,et al.  Causal Inference and Uplift Modelling: A Review of the Literature , 2017, PAPIs.

[107]  H. Chipman,et al.  BART: Bayesian Additive Regression Trees , 2008, 0806.3286.

[108]  Yao Zhang,et al.  Robust Recursive Partitioning for Heterogeneous Treatment Effects with Uncertainty Quantification , 2020, NeurIPS.

[109]  Chris S. Hulleman,et al.  Using Design Thinking to Improve Psychological Interventions: The Case of the Growth Mindset During the Transition to High School. , 2016, Journal of educational psychology.

[110]  H. Chipman,et al.  Bayesian Additive Regression Trees , 2006 .

[111]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[112]  Pieter Abbeel,et al.  Transfer Learning for Estimating Causal Effects using Neural Networks , 2018, ArXiv.