mlsanityData to 7 Oct 2026

Decision-Aware Proximal Bridge Learning for Optimal Treatment Selection

arXiv 2605.16989 ↗First posted 16 May 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Individualized treatment selection with continuous actions requires accurate causal response estimation in decision-relevant regions, rather than uniformly over the entire action space. Estimating a global causal response surface and then choosing the treatment that maximizes it can therefore be suboptimal, since standard estimation objectives allocate modeling effort according to the observed treatment distribution rather than the regions that determine the optimal decision. While decision-aware approaches have been studied in unconfounded settings, this problem remains underexplored in proximal causal inference, where proxy variables and bridge functions enable identification under suitable assumptions even in the presence of hidden confounding. Despite recent progress, proximal methods have primarily focused on treatment-effect and potential-outcome estimation rather than treatment selection and optimal decision-making. To bridge this gap, we introduce a policy-targeted weighted bridge loss that emphasizes decision-relevant treatment regions while retaining global stabilization. We prove a regret bound showing that the proposed weighted bridge loss controls treatment-selection regret through a weighted ill-posedness constant. We instantiate the framework in decision-aware variants of several proximal bridge solvers, yielding practical algorithms that alternate between weighted bridge estimation, response-surface projection, policy update, and weight refinement. Empirically, we find that decision-aware weighting reduces regret across several bridge solvers, suggesting improved treatment selection in proximal settings.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 47 references that are in the map. First posted on arXiv 16 May 2026.

Builds on

33 of 47 references.

References · 33

  1. 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,455
  2. 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 597
  3. 2017Smart "Predict, then Optimize"cited by 332
  4. 2016Learning Representations for Counterfactual Inferencecited by 327
  5. 2020Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representationscited by 154
  6. 2016Identifying Causal Effects With Proxy Variables of an Unmeasured Confoundercited by 138
  7. 2023Decision-Focused Learning: Foundations, State of the Art, Benchmark and Future Opportunitiescited by 129
  8. 2022Causal Transformer for Estimating Counterfactual Outcomescited by 121
  9. 2020Minimax Estimation of Conditional Moment Modelscited by 83
  10. 2020Estimating the Effects of Continuous-valued Interventions using Generative Adversarial Networkscited by 78
  11. 2020Semiparametric proximal causal inferencecited by 58
  12. 2018Policy Evaluation and Optimization with Continuous Treatmentscited by 57
21 more
  1. 2021Proximal Causal Learning with Kernels: Two-Stage Estimation and Moment Restrictioncited by 54
  2. 2021Causal Inference Under Unmeasured Confounding With Negative Controls: A Minimax Learning Approachcited by 43
  3. 2021Deep Proxy Causal Learning and its Application to Confounded Bandit Policy Evaluationcited by 39
  4. 2018A confounding bridge approach for double negative control inference on causal effectscited by 36
  5. 2020Kernel Methods for Causal Functions: Dose, Heterogeneous, and Incremental Response Curvescited by 36
  6. 2021Causal Decision Making and Causal Effect Estimation Are Not the Same... and Why It Matterscited by 25
  7. 2021Proximal Learning for Individualized Treatment Regimes Under Unmeasured Confoundingcited by 21
  8. 2021Minimax Kernel Machine Learning for a Class of Doubly Robust Functionals with Application to Proximal Causal Inferencecited by 20
  9. 2020Identifying effects of multiple treatments in the presence of unmeasured confoundingcited by 16
  10. 2021Proximal Causal Inference for Complex Longitudinal Studiescited by 16
  11. 2022Deep Learning Methods for Proximal Inference via Maximum Moment Restrictioncited by 13
  12. 2024IGC-Net for conditional average potential outcome estimation over timecited by 10
  13. 2025DeCaFlow: A deconfounding causal generative modelcited by 8
  14. 2025Practical do-Shapley Explanations with Estimand-Agnostic Causal Inferencecited by 6
  15. 2022Optimal Treatment Regimes for Proximal Causal Learningcited by 5
  16. 2023Doubly Robust Proximal Causal Learning for Continuous Treatmentscited by 4
  17. 2023Proximal Causal Inference without Uniqueness Assumptionscited by 3
  18. 2025Density Ratio-based Proxy Causal Learning Without Density Ratioscited by 3
  19. 2026Exactly Computing do-Shapley Valuescited by 3
  20. 2025Density Ratio-Free Doubly Robust Proxy Causal Learningcited by 1
  21. 2026CEPAE: Conditional Entropy-Penalized Autoencoders for Time Series Counterfactualscited by 1

Built on by

Not cited yet.