mlsanityData to 14 Oct 2026

Consistency of Neural Causal Partial Identification

arXiv 2405.15673 ↗First posted 24 May 2024cs.LG cs.AI stat.MLcited by 5

Not on the map. Why?

Abstract (arXiv)

Recent progress in Neural Causal Models (NCMs) showcased how identification and partial identification of causal effects can be automatically carried out via training of neural generative models that respect the constraints encoded in a given causal graph [Xia et al. 2022, Balazadeh et al. 2022]. However, formal consistency of these methods has only been proven for the case of discrete variables or only for linear causal models. In this work, we prove the consistency of partial identification via NCMs in a general setting with both continuous and categorical variables. Further, our results highlight the impact of the design of the underlying neural network architecture in terms of depth and connectivity as well as the importance of applying Lipschitz regularization in the training phase. In particular, we provide a counterexample showing that without Lipschitz regularization this method may not be asymptotically consistent. Our results are enabled by new results on the approximability of Structural Causal Models (SCMs) via neural generative models, together with an analysis of the sample complexity of the resulting architectures and how that translates into an error in the constrained optimization problem that defines the partial identification bounds.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 30 of its 55 references that are in the map. First posted on arXiv 24 May 2024.

Builds on

30 of 55 references.

References · 30

  1. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  2. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  3. 2018Lipschitz regularity of deep neural networks: analysis and efficient estimationcited by 415
  4. 2018Regularisation of Neural Networks by Enforcing Lipschitz Continuitycited by 314
  5. 2017Nearly-tight VC-dimension and pseudodimension bounds for piecewise linear neural networkscited by 311
  6. 2017Parseval Networks: Improving Robustness to Adversarial Examplescited by 293
  7. 2018Optimal approximation of continuous functions by very deep ReLU networkscited by 166
  8. 2017CausalGAN: Learning Causal Implicit Generative Models with Adversarial Trainingcited by 152
  9. 2020Training robust neural networks using Lipschitz boundscited by 99
  10. 2021The Causal-Neural Connection: Expressiveness, Learnability, and Inferencecited by 98
  11. 2020Faster Wasserstein Distance Estimation with the Sinkhorn Divergencecited by 73
  12. 2013Counterfactual Probabilities: Computational Methods, Bounds and Applicationscited by 67
18 more
  1. 2021Partial Counterfactual Identification from Observational and Experimental Datacited by 64
  2. 2022Neural Causal Models for Counterfactual Identification and Estimationcited by 57
  3. 2018Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distancecited by 56
  4. 2020Generative Interventions for Causal Learningcited by 45
  5. 2021An Automated Approach to Causal Inference in Discrete Settingscited by 42
  6. 2021CLIP: Cheap Lipschitz Training of Neural Networkscited by 29
  7. 2022Estimating average causal effects from patient trajectoriescited by 29
  8. 2020A Class of Algorithms for General Instrumental Variable Modelscited by 28
  9. 2022Stochastic Causal Programming for Bounding Treatment Effectscited by 25
  10. 2020A General Method for Deriving Tight Symbolic Bounds on Causal Effectscited by 18
  11. 2022Partial Identification of Treatment Effects with Implicit Generative Modelscited by 17
  12. 2023A Neural Framework for Generalized Causal Sensitivity Analysiscited by 17
  13. 2023Partial Counterfactual Identification of Continuous Outcomes with a Curvature Sensitivity Modelcited by 14
  14. 2015Nonparametric Bounds and Sensitivity Analysis of Treatment Effectscited by 12
  15. 2013Instrumentality Tests Revisitedcited by 8
  16. 2023Towards Bounding Causal Effects under Markov Equivalencecited by 8
  17. 2021High-Dimensional Distribution Generation Through Deep Neural Networkscited by 5
  18. 2015On Partial Identification of the Pure Direct Effectcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Information-Theoretic Causal Bounds under Unmeasured Confoundingcited by 2
  2. 2026Computational Identifiabilitycited by 2
  3. 2026Counterfactual identifiability beyond global monotonicity: non-monotone triangular structural causal modelscited by 0
  4. 2026Representable but Unlearned: Encoding Rank and the Interaction-Prediction Floorcited by 0
  5. 2026Counterfactual Rollouts for Offline Contrastive Goal-Conditioned Reinforcement Learningcited by 0