mlsanityData to 14 Oct 2026

Partial Identification of Treatment Effects with Implicit Generative Models

arXiv 2210.08139 ↗First posted 14 Oct 2022cs.LGcited by 17

Not on the map. Why?

Abstract (arXiv)

We consider the problem of partial identification, the estimation of bounds on the treatment effects from observational data. Although studied using discrete treatment variables or in specific causal graphs (e.g., instrumental variables), partial identification has been recently explored using tools from deep generative modeling. We propose a new method for partial identification of average treatment effects(ATEs) in general causal graphs using implicit generative models comprising continuous and discrete random variables. Since ATE with continuous treatment is generally non-regular, we leverage the partial derivatives of response functions to define a regular approximation of ATE, a quantity we call uniform average treatment derivative (UATD). We prove that our algorithm converges to tight bounds on ATE in linear structural causal models (SCMs). For nonlinear SCMs, we empirically show that using UATD leads to tighter and more stable bounds than methods that directly optimize the ATE.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 18 of its 56 references that are in the map. First posted on arXiv 14 Oct 2022.

Builds on

18 of 56 references.

References · 18

  1. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  2. 2017Learning Generative Models with Sinkhorn Divergencescited by 314
  3. 2015Data-driven Distributionally Robust Optimization Using the Wasserstein Metric: Performance Guarantees and Tractable Reformulationscited by 291
  4. 2017CausalGAN: Learning Causal Implicit Generative Models with Adversarial Trainingcited by 152
  5. 2021Counterfactual Generative Networkscited by 111
  6. 2021The Causal-Neural Connection: Expressiveness, Learnability, and Inferencecited by 98
  7. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  8. 2015Nonparametric methods for doubly robust estimation of continuous treatment effectscited by 64
  9. 2021Partial Counterfactual Identification from Observational and Experimental Datacited by 64
  10. 2021An Automated Approach to Causal Inference in Discrete Settingscited by 42
  11. 2020A Class of Algorithms for General Instrumental Variable Modelscited by 28
  12. 2022Stochastic Causal Programming for Bounding Treatment Effectscited by 25
6 more
  1. 2017Causal Generative Neural Networkscited by 24
  2. 2015Nonparametric Bounds and Sensitivity Analysis of Treatment Effectscited by 12
  3. 2022Partial Identification with Noisy Covariates: A Robust Optimization Approachcited by 11
  4. 2020Partial Identifiability in Discrete Data With Measurement Errorcited by 10
  5. 2019Incremental causal effectscited by 7
  6. 2015On Partial Identification of the Pure Direct Effectcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 7; 2024: 6; 2025: 0; 2026: 4.

17 papers.

Sort:

  1. 2023Sharp Bounds for Generalized Causal Sensitivity Analysiscited by 24
  2. 2023Counterfactual Generative Models for Time-Varying Treatmentscited by 17
  3. 2023Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Modelcited by 14
  4. 2023Structured Neural Networks for Density Estimation and Causal Inferencecited by 13
  5. 2024Meta-Learners for Partially-Identified Treatment Effects Across Multiple Environmentscited by 11
  6. 2024Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inferencecited by 9
  7. 2023Towards Bounding Causal Effects under Markov Equivalencecited by 8
  8. 2024Causality from Bottom to Top: A Surveycited by 5
  9. 2024Consistency of Neural Causal Partial Identificationcited by 5
  10. 2024Conditional Generative Models are Sufficient to Sample from Any Causal Effect Estimandcited by 4
  11. 2024Learning Representations of Instruments for Partial Identification of Treatment Effectscited by 4
  12. 2023Approximate Causal Effect Identification under Weak Confoundingcited by 3
5 more
  1. 2026IV-ICL: Bounding Causal Effects with Instrumental Variables via In-Context Learningcited by 3
  2. 2023CTP:A Causal Interpretable Model for Non-Communicable Disease Progression Predictioncited by 2
  3. 2026Information-Theoretic Causal Bounds under Unmeasured Confoundingcited by 2
  4. 2026Foundation Models for Partial Causal Identificationcited by 2
  5. 2026Uncertainty-aware Causal Decision Making via Effect Bound Decompositioncited by 0