mlsanityData to 7 Oct 2026

Unpaired Multi-Domain Causal Representation Learning

arXiv 2302.00993 ↗First posted 2 Feb 2023stat.ML cs.LG stat.MEcited by 32

Not on the map. Why?

Abstract (arXiv)

The goal of causal representation learning is to find a representation of data that consists of causally related latent variables. We consider a setup where one has access to data from multiple domains that potentially share a causal representation. Crucially, observations in different domains are assumed to be unpaired, that is, we only observe the marginal distribution in each domain but not their joint distribution. In this paper, we give sufficient conditions for identifiability of the joint distribution and the shared causal graph in a linear setup. Identifiability holds if we can uniquely recover the joint distribution and the shared causal representation from the marginal distributions in each domain. We transform our identifiability results into a practical method to recover the shared latent causal graph.

Key facts. Cited by 32 papers in the map (data through 7 Oct 2026). Builds on 21 of its 76 references that are in the map. First posted on arXiv 2 Feb 2023.

Builds on

21 of 76 references.

References · 21

  1. 2019A Comprehensive Survey on Transfer Learningcited by 1,154
  2. 2018Recognition in Terra Incognitacited by 655
  3. 2019Variational Autoencoders and Nonlinear ICA: A Unifying Frameworkcited by 494
  4. 2021Self-Supervised Representation Learning: Introduction, Advances and Challengescited by 147
  5. 2022Interventional Causal Representation Learningcited by 140
  6. 2020On Linear Identifiability of Learned Representationscited by 103
  7. 2022Linear Causal Disentanglement via Interventionscited by 84
  8. 2022Function Classes for Identifiable Nonlinear Independent Component Analysiscited by 53
  9. 2022Latent Hierarchical Causal Structure Discovery with Rank Constraintscited by 50
  10. 2022Identifying Weight-Variant Latent Causal Modelscited by 33
  11. 2022Learning Causal Representations of Single Cells via Sparse Mechanism Shift Modelingcited by 33
  12. 2018High-Dimensional Causal Discovery Under non-Gaussianitycited by 24
9 more
  1. 2014Group Factor Analysiscited by 15
  2. 2018MAGAN: Aligning Biological Manifoldscited by 15
  3. 2020Causal Discovery with Multi-Domain LiNGAM for Latent Factorscited by 13
  4. 2022Independence Testing-Based Approach to Causal Discovery under Measurement Error and Linear Non-Gaussian Modelscited by 13
  5. 2013Independent Vector Analysis: Identification Conditions and Performance Boundscited by 7
  6. 2021Shared Independent Component Analysis for Multi-Subject Neuroimagingcited by 6
  7. 2022Multi-View Independent Component Analysis with Shared and Individual Sourcescited by 6
  8. 2019Multi-Domain Translation by Learning Uncoupled Autoencoderscited by 3
  9. 2022Non-Independent Components Analysiscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 13; 2025: 6; 2026: 10.

32 papers.

Sort:

  1. 2024A Sparsity Principle for Partially Observable Causal Representation Learningcited by 34
  2. 2023Causal Representation Learning Made Identifiable by Grouping of Observational Variablescited by 30
  3. 2024Marrying Causal Representation Learning with Dynamical Systems for Sciencecited by 22
  4. 2023From Identifiable Causal Representations to Controllable Counterfactual Generation: A Survey on Causal Generative Modelingcited by 21
  5. 2024Linear Causal Representation Learning from Unknown Multi-node Interventionscited by 14
  6. 2024Causal Representation Learning from Multimodal Biomedical Observationscited by 12
  7. 2024Identifiability Guarantees for Causal Disentanglement from Purely Observational Datacited by 7
  8. 2024Propensity Score Alignment of Unpaired Multimodal Datacited by 6
  9. 2024Toward the Identifiability of Comparative Deep Generative Modelscited by 4
  10. 2024Identifying General Mechanism Shifts in Linear Causal Representationscited by 4
  11. 2026Distributional Equivalence in Linear Non-Gaussian Latent-Variable Cyclic Causal Models: Characterization and Learningcited by 4
  12. 2023Conditionally Invariant Representation Learning for Disentangling Cellular Heterogeneitycited by 3
20 more
  1. 2024Identifiable Shared Component Analysis of Unpaired Multimodal Mixturescited by 3
  2. 2024Differentiable Causal Discovery For Latent Hierarchical Causal Modelscited by 3
  3. 2025Causal Structure and Representation Learning with Biomedical Applicationscited by 3
  4. 2024Contrastive independent component analysiscited by 2
  5. 2024Content-Style Learning from Unaligned Domains: Identifiability under Unknown Latent Dimensionscited by 2
  6. 2026Score-Based Causal Discovery of Latent Variable Causal Modelscited by 2
  7. 2024Unsupervised discovery of the shared and private geometry in multi-view datacited by 1
  8. 2025Intervening to Learn and Compose Causally Disentangled Representationscited by 1
  9. 2025Score-based Greedy Search for Structure Identification of Partially Observed Linear Causal Modelscited by 1
  10. 2026Nonlinear multi-study sparse factor analysiscited by 1
  11. 2026Content-Style Identification via Differential Independencecited by 1
  12. 2026MultiLoReFT: Decoupling Shared and Modality-Specific Subspaces in Multimodal Learning via Low-Rank Representation Fine-Tuningcited by 1
  13. 2025Multi-View Causal Discovery without Non-Gaussianity: Identifiability and Algorithmscited by 0
  14. 2025Trek-Based Parameter Identification for Linear Causal Models With Arbitrarily Structured Latent Variablescited by 0
  15. 2025PersonaX: Multimodal Datasets with LLM-Inferred Behavior Traitscited by 0
  16. 2026Provable Subspace Identification of Nonlinear Multi-view CCAcited by 0
  17. 2026Identifiable Multimodal Causal Representation Learning under Partial Latent Sharingcited by 0
  18. 2026FiGuRO: Intrinsic Dimension Estimation for Multi-Modal Datacited by 0
  19. 2026Jigsaw-CRL: Recovering Global Latent Causal Order from Fragmented Multi-Client Interventionscited by 0
  20. 2026Hessian Rank Constraint for Learning Structure of Nonlinear Latent Variable Modelscited by 0