{"type":"ScholarlyArticle","title":"Estimating Conditional Average Treatment Effects via Sufficient Representation Learning","authors":[{"name":"Pengfei Shi","slug":"pengfei-shi","given":"Pengfei","family":"Shi","affiliation":"Xiamen University , Xiamen , China; Ant Group , Hangzhou , China"},{"name":"Wei Zhong","slug":"wei-zhong","given":"Wei","family":"Zhong","affiliation":"Xiamen University , Xiamen , China"},{"name":"Xinyu Zhang","slug":"xinyu-zhang","given":"Xinyu","family":"Zhang","affiliation":"Academy of Mathematics and Systems Science , Chinese Academy of Sciences , Beijing , China"},{"name":"Ningtao Wang","slug":"ningtao-wang","given":"Ningtao","family":"Wang","affiliation":"Ant Group , Hangzhou , China"},{"name":"Xing Fu","slug":"xing-fu","given":"Xing","family":"Fu","affiliation":"Ant Group , Hangzhou , China"},{"name":"Weiqiang Wang","slug":"weiqiang-wang","given":"Weiqiang","family":"Wang","affiliation":"Ant Group , Hangzhou , China"},{"name":"Yin Jin","slug":"yin-jin","given":"Yin","family":"Jin","affiliation":"Ant Group , Hangzhou , China"}],"abstract":"Estimating the conditional average treatment effects (CATE) is very important in causal inference and has a wide range of applications across many fields. In the estimation process of CATE, the unconfoundedness assumption is typically required to ensure the identifiability of the regression problems. When estimating CATE using high-dimensional data, there have been many variable selection methods and neural network approaches based on representation learning, while these methods do not provide a way to verify whether the subset of variables after dimensionality reduction or the learned representations still satisfy the unconfoundedness assumption during the estimation process, which can lead to ineffective estimates of the treatment effects. Additionally, these methods typically use data from only the treatment or control group when estimating the regression functions for each group. This paper proposes a novel neural network approach named CrossNet to learn a sufficient representation for the features, based on which we then estimate the CATE, where cross indicates that in estimating the regression functions, we used data from their own group as well as cross-utilized data from another group. Numerical simulations and empirical results demonstrate that our method outperforms the competitive approaches.","doi":"10.24963/ijcai.2024/541","url":"https://ijcai.org/proceedings/2024/541","pdf":"https://ijcai.org/proceedings/2024/0541.pdf","markdown":"https://ijcai.org/proceedings/2024/541.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/541","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Machine Learning","track":"Main Track","number":541,"pages":{"first":4894,"last":4901},"keywords":["Machine Learning: ML: Causality","Machine Learning: ML: Regression","Machine Learning: ML: Supervised Learning","Machine Learning: ML: Representation learning"],"videos":[],"references":[{"position":1,"raw":"[Athey and Imbens, 2016] Susan Athey and Guido Imbens. Recursive partitioning for heterogeneous causal effects. Proceedings of the National Academy of Sciences, 113(27):7353-7360, 2016.","title":"Recursive partitioning for heterogeneous causal effects","authors":"Imbens Athey, Susan Athey, Guido Imbens","year":"2016","venue":"Proceedings of the National Academy of Sciences","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chauhan et al., 2023] Vinod K Chauhan, Soheila Molaei, Marzia Hoque Tania, Anshul Thakur, Tingting Zhu, and David A Clifton. Adversarial de-confounding in individu- alised treatment effects estimation. In International Con- ference on Artificial Intelligence and Statistics, pages 837- 849. PMLR, 2023.","title":"Adversarial de-confounding in individualised treatment effects estimation","authors":"Soheila Vinod K Chauhan, Marzia Molaei, Anshul Hoque Tania, Tingting Thakur, David A Zhu, Clifton","year":"2023","venue":"International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Curth and van der Schaar, 2021] Alicia Curth and Mihaela van der Schaar. Nonparametric estimation of heteroge- neous treatment effects: From theory to learning algo- rithms. In International Conference on Artificial Intelli- gence and Statistics, pages 1810-1818. PMLR, 2021.","title":"Nonparametric estimation of heterogeneous treatment effects: From theory to learning algorithms","authors":"Van Curth, Alicia Der Schaar, Mihaela Curth, Van Der Schaar","year":"2021","venue":"International Conference on Artificial Intelligence and Statistics","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Green and Kern, 2012] Donald P Green and Holger L Kern. Modeling heterogeneous treatment effects in survey ex- periments with bayesian additive regression trees. Public opinion quarterly, 76(3):491-511, 2012.","title":"Modeling heterogeneous treatment effects in survey experiments with bayesian additive regression trees","authors":"P Donald, Green, L Holger, Kern","year":"2012","venue":"Public opinion quarterly","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Hansen and Bowers, 2009] Ben B Hansen and Jake Bow- ers. Attributing effects to a cluster-randomized get-out- the-vote campaign. Journal of the American Statistical Association, 104(487):873-885, 2009.","title":"Attributing effects to a cluster-randomized get-outthe-vote campaign","authors":"B Ben, Jake Hansen, Bowers","year":"2009","venue":"Journal of the American Statistical Association","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Hassanpour and Greiner, 2019a] Negar Hassanpour and Russell Greiner. Counterfactual regression with impor- tance sampling weights. In IJCAI, pages 5880-5887, 2019.","title":"Counterfactual regression with importance sampling weights","authors":"Greiner Hassanpour, Negar Hassanpour, Russell Greiner","year":"2019","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2019/815"},{"position":7,"raw":"[Hassanpour and Greiner, 2019b] Negar Hassanpour and Russell Greiner. Learning disentangled representations for counterfactual regression. In International Conference on Learning Representations, 2019.","title":"Learning disentangled representations for counterfactual regression","authors":"Greiner Hassanpour, Negar Hassanpour, Russell Greiner","year":"2019","venue":"International Conference on Learning Representations","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Hill, 2011] Jennifer L Hill. Bayesian nonparametric mod- eling for causal inference. Journal of Computational and Graphical Statistics, 20(1):217-240, 2011.","title":"Bayesian nonparametric modeling for causal inference","authors":"L Jennifer, Hill","year":"2011","venue":"Journal of Computational and Graphical Statistics","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Kennedy, 2020] Edward H Kennedy. Towards optimal dou- bly robust estimation of heterogeneous causal effects. arXiv preprint arXiv:2004.14497, 2020.","title":"Towards optimal doubly robust estimation of heterogeneous causal effects","authors":"H Edward, Kennedy","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Kulis et al., 2009] Brian Kulis, Mátyás A Sustik, and Inder- jit S Dhillon. Low-rank kernel learning with bregman ma- trix divergences. Journal of Machine Learning Research, 10(2), 2009.","title":"Low-rank kernel learning with bregman matrix divergences","authors":"Brian Kulis, Mátyás A Sustik, Inderjit S Dhillon","year":"2009","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Künzel et al., 2019] Sören R Künzel, Jasjeet S Sekhon, Pe- ter J Bickel, and Bin Yu. Metalearners for estimat- ing heterogeneous treatment effects using machine learn- ing. Proceedings of the national academy of sciences, 116(10):4156-4165, 2019.","title":"Metalearners for estimating heterogeneous treatment effects using machine learning","authors":"Sören R Künzel, Peter J Jasjeet S Sekhon, Bin Bickel, Yu","year":"2019","venue":"Proceedings of the national academy of sciences","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[LaLonde, 1986] Robert J LaLonde. Evaluating the econo- metric evaluations of training programs with experimen- tal data. The American economic review, pages 604-620, 1986.","title":"Evaluating the econometric evaluations of training programs with experimental data","authors":"J Robert, Lalonde","year":"1986","venue":"The American economic review","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Mooij et al., 2016] Joris M Mooij, Jonas Peters, Dominik Janzing, Jakob Zscheischler, and Bernhard Schölkopf. Distinguishing cause from effect using observational data: methods and benchmarks. The Journal of Machine Learn- ing Research, 17(1):1103-1204, 2016.","title":"Distinguishing cause from effect using observational data: methods and benchmarks","authors":"Jonas Joris M Mooij, Dominik Peters, Jakob Janzing, Bernhard Zscheischler, Schölkopf","year":"2016","venue":"The Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Robins and Rotnitzky, 1995] James M Robins and Andrea Rotnitzky. Semiparametric efficiency in multivariate re- gression models with missing data. Journal of the Ameri- can Statistical Association, 90(429):122-129, 1995.","title":"Semiparametric efficiency in multivariate regression models with missing data","authors":"M James, Andrea Robins, Rotnitzky","year":"1995","venue":"Journal of the American Statistical Association","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Rolling and Yang, 2014] Craig A Rolling and Yuhong Yang. Model selection for estimating treatment effects. Journal of the Royal Statistical Society: Series B (Statisti- cal Methodology), 76(4):749-769, 2014.","title":"Model selection for estimating treatment effects","authors":"A Craig, Yuhong Rolling, Yang","year":"2014","venue":"Journal of the Royal Statistical Society: Series B (Statistical Methodology","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Rosenbaum, 2007] Paul R Rosenbaum. Interference be- tween units in randomized experiments. Journal of the american statistical association, 102(477):191-200, 2007.","title":"Interference between units in randomized experiments","authors":"Paul R Rosenbaum","year":"2007","venue":"Journal of the american statistical association","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Rubin, 2005] Donald B Rubin. Causal inference using po- tential outcomes: Design, modeling, decisions. Journal of the American Statistical Association, 100(469):322-331, 2005.","title":"Causal inference using potential outcomes: Design, modeling, decisions","authors":"Donald B Rubin","year":"2005","venue":"Journal of the American Statistical Association","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Shalit et al., 2017] Uri Shalit, Fredrik D Johansson, and David Sontag. Estimating individual treatment effect: gen- eralization bounds and algorithms. In International con- ference on machine learning, pages 3076-3085. PMLR, 2017.","title":"Estimating individual treatment effect: generalization bounds and algorithms","authors":"Shalit","year":"2017","venue":"International conference on machine learning","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Shi et al., 2019] Claudia Shi, David Blei, and Victor Veitch. Adapting neural networks for the estimation of treatment effects. Advances in neural information processing sys- tems, 32, 2019.","title":"Adapting neural networks for the estimation of treatment effects","authors":"Claudia Shi, David Blei, Victor Veitch","year":"2019","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Smith and Todd, 2005] Jeffrey A Smith and Petra E Todd. Does matching overcome lalonde's critique of nonexper- imental estimators? Journal of econometrics, 125(1- 2):305-353, 2005.","title":"Does matching overcome lalonde's critique of nonexperimental estimators?","authors":"A Jeffrey, Petra E Smith, Todd","year":"2005","venue":"Journal of econometrics","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[VanderWeele, 2019] Tyler J VanderWeele. Principles of confounder selection. European journal of epidemiology, 34:211-219, 2019.","title":"Principles of confounder selection","authors":"J Tyler, Vanderweele","year":"2019","venue":"European journal of epidemiology","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Wager and Athey, 2018] Stefan Wager and Susan Athey. Estimation and inference of heterogeneous treatment ef- fects using random forests. Journal of the American Sta- tistical Association, 113(523):1228-1242, 2018.","title":"Estimation and inference of heterogeneous treatment effects using random forests","authors":"Athey Wager, Stefan Wager, Susan Athey","year":"2018","venue":"Journal of the American Statistical Association","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Whitmore, 2005] Diane Whitmore. Resource and peer im- pacts on girls' academic achievement: Evidence from a randomized experiment. American Economic Review, 95(2):199-203, 2005.","title":"Resource and peer impacts on girls' academic achievement: Evidence from a randomized experiment","authors":"Diane Whitmore","year":"2005","venue":"American Economic Review","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Yao et al., 2021] Liuyi Yao, Zhixuan Chu, Sheng Li, Yaliang Li, Jing Gao, and Aidong Zhang. A survey on causal inference. ACM Transactions on Knowledge Dis- covery from Data (TKDD), 15(5):1-46, 2021.","title":"A survey on causal inference","authors":"Yao","year":"2021","venue":"ACM Transactions on Knowledge Discovery from Data (TKDD)","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yu et al., 2021] Shujian Yu, Ammar Shaker, Francesco Ale- siani, and Jose Principe. Measuring the discrepancy be- tween conditional distributions: methods, properties and applications. In Proceedings of the Twenty-Ninth Interna- tional Conference on International Joint Conferences on Artificial Intelligence, pages 2777-2784, 2021.","title":"Measuring the discrepancy between conditional distributions: methods, properties and applications","authors":"Yu","year":"2021","venue":"Proceedings of the Twenty-Ninth International Conference on International Joint Conferences on Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/2020/385"},{"position":26,"raw":"[Zhang et al., 2021] Weijia Zhang, Lin Liu, and Jiuyong Li. Treatment effect estimation with disentangled latent fac- tors. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 10923-10930, 2021.","title":"Treatment effect estimation with disentangled latent factors","authors":"Weijia Zhang, Lin Liu, Jiuyong Li","year":"2021","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}