mlsanityData to 13 Oct 2026

Estimating Average Treatment Effects via Orthogonal Regularization

arXiv 2101.08490 ↗First posted 21 Jan 2021cs.LG stat.MLcited by 25

Not on the map. Why?

Abstract (arXiv)

Decision-making often requires accurate estimation of treatment effects from observational data. This is challenging as outcomes of alternative decisions are not observed and have to be estimated. Previous methods estimate outcomes based on unconfoundedness but neglect any constraints that unconfoundedness imposes on the outcomes. In this paper, we propose a novel regularization framework for estimating average treatment effects that exploits unconfoundedness. To this end, we formalize unconfoundedness as an orthogonality constraint, which ensures that the outcomes are orthogonal to the treatment assignment. This orthogonality constraint is then included in the loss function via a regularization. Based on our regularization framework, we develop deep orthogonal networks for unconfounded treatments (DONUT), which learn outcomes that are orthogonal to the treatment assignment. Using a variety of benchmark datasets for estimating average treatment effects, we demonstrate that DONUT outperforms the state-of-the-art substantially.

Key facts. Cited by 25 papers in the map (data through 13 Oct 2026). Builds on 16 of its 70 references that are in the map. First posted on arXiv 21 Jan 2021.

Builds on

16 of 70 references.

References · 16

  1. 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 598
  2. 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 409
  3. 2017Quasi-Oracle Estimation of Heterogeneous Treatment Effectscited by 330
  4. 2016Learning Representations for Counterfactual Inferencecited by 327
  5. 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 285
  6. 2016Optimal and Adaptive Off-policy Evaluation in Contextual Banditscited by 111
  7. 2017Balanced Policy Evaluation and Learningcited by 51
  8. 2016Approximate Residual Balancing: De-Biased Inference of Average Treatment Effects in High Dimensionscited by 50
  9. 2018Benchmarking Framework for Performance-Evaluation of Causal Inference Analysiscited by 33
  10. 2020Matching in Selective and Balanced Representation Space for Treatment Effects Estimationcited by 30
  11. 2016Recursive Partitioning for Personalization using Observational Datacited by 28
  12. 2020Estimating Individual Treatment Effects with Time-Varying Confounderscited by 21
4 more
  1. 2018Linked Causal Variational Autoencoder for Inferring Paired Spillover Effectscited by 16
  2. 2015Structural Nested Models and G-estimation: The Partially Realized Promisecited by 15
  3. 2019Debiased Bayesian inference for average treatment effectscited by 7
  4. 2017Estimating Average Treatment Effects: Supplementary Analyses and Remaining Challengescited by 5

Built on by

Cited by year

Citing papers per year: 2021: 6; 2022: 8; 2023: 3; 2024: 6; 2025: 2.

25 papers.

Sort:

By area:

  1. 2021On Inductive Biases for Heterogeneous Treatment Effect Estimationcited by 74
  2. 2022Causal Machine Learning for Healthcare and Precision Medicinecited by 68
  3. 2021Causal Effect Inference for Structured Treatmentscited by 38
  4. 2022Combining Observational and Randomized Data for Estimating Heterogeneous Treatment Effectscited by 32
  5. 2021Sequential Deconfounding for Causal Inference with Unobserved Confounderscited by 24
  6. 2022Normalizing Flows for Interventional Density Estimationcited by 21
  7. 2024DiffPO: A causal diffusion model for learning distributions of potential outcomescited by 17
  8. 2021Generalizing Off-Policy Learning under Sample Selection Biascited by 16
  9. 2021Deconfounding Temporal Autoencoder: Estimating Treatment Effects over Time Using Noisy Proxiescited by 16
  10. 2021AttDMM: An Attentive Deep Markov Model for Risk Scoring in Intensive Care Unitscited by 14
  11. 2023Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Modelcited by 14
  12. 2022Deep Causal Learning: Representation, Discovery and Inferencecited by 11
13 more
  1. 2022A Survey of Deep Causal Models and Their Industrial Applicationscited by 10
  2. 2022Estimating Conditional Average Treatment Effects with Missing Treatment Informationcited by 7
  3. 2023Estimating Causal Effects using a Multi-task Deep Ensemblecited by 7
  4. 2024Stabilized Neural Prediction of Potential Outcomes in Continuous Timecited by 7
  5. 2024Causal Machine Learning for Cost-Effective Allocation of Development Aidcited by 6
  6. 2024Constructing Confidence Intervals for Average Treatment Effects from Multiple Datasetscited by 6
  7. 2024Causal-StoNet: Causal Inference for High-Dimensional Complex Datacited by 3
  8. 2022Detecting User Exits from Online Behavior: A Duration-Dependent Latent State Modelcited by 2
  9. 2022Moderately-Balanced Representation Learning for Treatment Effects with Orthogonality Informationcited by 1
  10. 2024A Multi-class Ride-hailing Service Subsidy System Utilizing Deep Causal Networkscited by 1
  11. 2023Multiple Instance Learning for Uplift Modelingcited by 0
  12. 2025TSCAN: Context-Aware Uplift Modeling via Two-Stage Training for Online Merchant Business Diagnosiscited by 0
  13. 2025Deconfounded Warm-Start Thompson Sampling with Applications to Precision Medicinecited by 0