Abstract
Counterfactual inference for continuous rather than binary treatment variables is more common in real-world causal inference tasks. While there are already some sample reweighting methods based on Marginal Structural Model for eliminating the confounding bias, they generally focus on removing the treatment’s linear dependence on confounders and rely on the accuracy of the assumed parametric models, which are usually unverifiable. In this paper, we propose a de-confounding representation learning (DRL) framework for counterfactual outcome estimation of continuous treatment by generating the representations of covariates decorrelated with the treatment variables. The DRL is a non-parametric model that eliminates both linear and nonlinear dependence between treatment and covariates. Specifically, we train the correlations between the de-confounding representations and the treatment variables against the correlations between the covariate representations and the treatment variables to eliminate confounding bias. Further, a counterfactual inference network is embedded into the framework to make the learned representations serve both de-confounding and trusted inference. Extensive experiments on synthetic and semi-synthetic datasets show that the DRL model performs superiorly in learning de-confounding representations and outperforms state-of-the-art counterfactual inference models for continuous treatment variables. In addition, we apply the DRL model to a real-world medical dataset MIMIC III and demonstrate a detailed causal relationship between red cell width distribution and mortality.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10618-024-01058-3/MediaObjects/10618_2024_1058_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10618-024-01058-3/MediaObjects/10618_2024_1058_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10618-024-01058-3/MediaObjects/10618_2024_1058_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs10618-024-01058-3/MediaObjects/10618_2024_1058_Fig4_HTML.png)
References
Austin Peter C (2011) An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res 46(3):399–424
Bellot A, Dhir A, Prando G (2023) Generalization bounds and algorithms for estimating the effect of multiple treatments and dosage
Bica I, Jordon J, Schaar M (2020) Estimating the effects of continuous-valued interventions using generative adversarial networks. Adv Neural Inf Process Syst 33:16434–16445
Brooks-Gunn J, Liaw F-R, Klebanov PK (1992) Effects of early intervention on cognitive function of low birth weight preterm infants. J Pediatr 120(3):350–359
Castro-Martín L, Mar Rueda M, Ferri-García R (2022) Combining statistical matching and propensity score adjustment for inference from non-probability surveys. J Comput Appl Math 404(2):113414
Chang Y, Dy J (2017) Informative subspace learning for counterfactual inference. In Proceedings of the AAAI Conference on Artificial Intelligence, pp 31
Chipman HA, George EI, McCulloch RE (2010) Bart: bayesian additive regression trees. Annal Appl Stat 4(1):69
DeStefano F (2007) Vaccines and autism: evidence does not support a causal association. Clin Pharmacol Therap 82(6):756–759
Du X, Sun L, Duivesteijn W, Nikolaev A, Pechenizkiy M (2021) Adversarial balancing-based representation learning for causal effect inference with observational data. Data Min Knowl Disc 35(4):1713–1738
D’Aunno T (2010) Reputation and power: organizational image and pharmaceutical regulation at the fda. Adm Sci Q 55(4):671–672
Fong C, Hazlett C, Imai K (2018) Covariate balancing propensity score for a continuous treatment: application to the efficacy of political advertisements. Annal Appl Stat 12(1):156–177
Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. Adv Neural Inf Process Syst 27:69
Horne BD, May HT, Kfoury AG, Renlund DG, Muhlestein JB, Lappé D, Rasmusson KD, Bunch TJ, Carlquist JF, Bair TL (2014) The intermountain risk score (including the red cell distribution width) predicts heart failure and other morbidity endpoints. Eur J Heart Fail 12(11):1203–1213
Hunziker S, Celi LA, Lee J, Howell MD (2012) Red cell distribution width improves the simplified acute physiology score for risk prediction in unselected critically ill patients. Crit Care 16:1–8
Imai K, Ratkovic M (2014) Covariate balancing propensity score. J R Stat Soc Ser B Stat Methodol 76(1):243–263
Imbens GW (2000) The role of the propensity score in estimating dose-response functions. Biometrika 87(3):706–710
Imbens GW (2004) Nonparametric estimation of average treatment effects under exogeneity: a review. Rev Econ Stat 86(1):4–29
Imbens GW, Rubin DB (2015) Causal inference for statistics, social, and biomedical sciences. Cambridge University Press, Cambridgeshire
Johansson FD, Kallus N, Shalit U, Sontag D (2018) Learning weighted representations for generalization across designs. ar**v e-prints ar**v:1802.08598
Johansson FD, Shalit U, Sontag D (2016) Learning representations for counterfactual inference. In Proceedings of The 33rd International Conference on Machine Learning 48:3020–3029
Johnson AE, Pollard TJ, Shen L, Lehman LH, Feng M, Ghassemi M, Moody B, Szolovits P, Anthony Celi L, Mark RG (2016) Mimic-iii, a freely accessible critical care database. Sci Data 3(1):1–9
Johnson A, Pollard T, Mark R (2016) Mimic-iii clinical database (version 1.4). PhysioNet
Kallus N (2020) Generalized optimal matching methods for causal inference. J Mach Learn Res 21(62):1–54
Kallus N, Santacatterina M (2019) Kernel optimal orthogonality weighting: A balancing approach to estimating effects of continuous treatments. ar** precision treatment rules with observational data. Behav Res Ther 120:103412
Kluve J, Schneider H, Uhlendorff A, Zhao Z (2012) Evaluating continuous training programmes by using the generalized propensity score. J R Stat Soc Ser A Stat Soc 175(2):587–617
Kohavi R, Longbotham R (2011) Unexpected results in online controlled experiments. ACM SIGKDD Explorations Newsl 12(2):31–35
Kreif N, Grieve R, Díaz I, Harrison D (2015) Evaluation of the effect of a continuous treatment: a machine learning approach with an application to treatment for traumatic brain injury. Health Econ 24(9):1213–1228
Lee JH, Chung HJ, Kim K, Jo YH, Rhee JE, Kim YJ, Kang KW (2013) Red cell distribution width as a prognostic marker in patients with community-acquired pneumonia. Am J Emerg Med 31(1):72–79
Lee BK, Lessler J, Stuart EA (2011) Weight trimming and propensity score weighting. PLoS ONE 6(3):18174
Li KKYLB, Cui P, Yang H, Tao J, Wu F (2021) continuous treatment effect estimation through generative adversaria1 de confounding. Data Min Knowl Disc 35(6):2467–2497
Ma X, Wang J (2019) Robust inference using inverse probability weighting*. J Am Stat Assoc 115(532):1–26
Malina D, Bothwell LE, Greene JA, Podolsky SH, Jones DS (2016) Assessing the gold standard — lessons from the history of rcts. N Engl J Med 374(22):2175–2181
Myers JA, Rassen JA, Gagne JJ, Huybrechts KF, Schneeweiss S, Rothman KJ, Joffe MM, Glynn RJ (2011) Effects of adjusting for instrumental variables on bias and precision of effect estimates. Am J Epidemiol 174(11):1213–1222
Nie L, Ye M, Liu Q, Nicolae D (2021) Vcnet and functional targeted regularization for learning causal effects of continuous treatments. ar**v e-prints ar**v:2103.07861
Nijsse M (1991) Multiple correlation-coefficient. Biometrics 47(1):341–341
Pearl J (2009) Causality. Cambridge University Press, Cambridgeshire
Robins JM, Hernán M, Brumback B (2000) Marginal structural models and causal inference in epidemiology. Epidemiology 11(5):550–560
Robins JM, Rotnitzky A, Zhao L (1994) Estimation of regression coefficients when some regressors are not always observed. J Am Stat Assoc 89(427):846–866
Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70(1):41–55
Rubin DB (1974) Estimating causal effects of treatments in randomized and nonrandomized studies. J Educ Psychol 66(5):688
Schwab P, Linhardt L, Karlen W (2018) Perfect match: A simple method for learning representations for counterfactual inference with neural networks. ar**v e-prints ar**v:1810.00656
Schwab P, Linhardt L, Bauer S, Buhmann JM, Karlen W (2020) Learning counterfactual representations for estimating individual dose-response curves. In Proceedings of the AAAI Conference on Artificial Intelligence 34(4):5612–5619
Yao L, Chu Z, Li S, Li Y, Gao J, Zhang A (2021) A survey on causal inference. ACM Trans Knowl Discov Data 15(5):1–46
Zhang Y-F, Zhang H, Lipton ZC, Li LE, **ng EP (2022) Exploring transformer backbones for heterogeneous treatment effect estimation. ar**v e-prints ar**v:2202.01336
Zhao Y, Huang Q, Fu S, Sun H (2023) Does misclassifying non-confounding covariates as confounders affect the causal inference within the potential outcomes framework? ar**v e-prints ar**v:2308.11676
Zhu Y, Coffman DL, Ghosh D (2015) A boosting algorithm for estimating generalized propensity scores with continuous treatments. J Causal Inference 3(1):25–40
Zou WY, Shyam S, Mui M, Wang M, Pedersen J, Ghahramani Z (2020) Learning continuous treatment policy and bipartite embeddings for matching with heterogeneous causal effects. ar**v e-prints ar**v:2004.09703
Funding
The authors thank funding support from the National Natural Science Foundation of China (No. 62372210), Natural Science Foundation of Jilin Province (No. 20240101025JJ).
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by YZ, QH, HZ, YP and HS. The first draft of the manuscript was written by YZ and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no Conflict of interest to declare that are relevant to the content of this article.
Additional information
Responsible editor: Matteo Riondato.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhao, Y., Huang, Q., Zeng, H. et al. De-confounding representation learning for counterfactual inference on continuous treatment via generative adversarial network. Data Min Knowl Disc (2024). https://doi.org/10.1007/s10618-024-01058-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10618-024-01058-3