mlsanityData to 7 Oct 2026

Identifying General Mechanism Shifts in Linear Causal Representations

arXiv 2410.24059 ↗First posted 31 Oct 2024cs.LG cs.AI stat.MLcited by 4

Not on the map. Why?

Abstract (arXiv)

We consider the linear causal representation learning setting where we observe a linear mixing of d unknown latent factors, which follow a linear structural causal model. Recent work has shown that it is possible to recover the latent factors as well as the underlying structural causal model over them, up to permutation and scaling, provided that we have at least d environments, each of which corresponds to perfect interventions on a single latent node (factor). After this powerful result, a key open problem faced by the community has been to relax these conditions: allow for coarser than perfect single-node interventions, and allow for fewer than d of them, since the number of latent factors d could be very large. In this work, we consider precisely such a setting, where we allow a smaller than d number of environments, and also allow for very coarse interventions that can very coarsely textitchange the entire causal graph over the latent factors. On the flip side, we relax what we wish to extract to simply the textitlist of nodes that have shifted between one or more environments. We provide a surprising identifiability result that it is indeed possible, under some very mild standard assumptions, to identify the set of shifted nodes. Our identifiability proof moreover is a constructive one: we explicitly provide necessary and sufficient conditions for a node to be a shifted node, and show that we can check these conditions given observed data. Our algorithm lends itself very naturally to the sample setting where instead of just interventional distributions, we are provided datasets of samples from each of these distributions. We corroborate our results on both synthetic experiments as well as an interesting psychometric dataset. The code can be found at https://github.com/TianyuCodings/iLCS.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 21 of its 50 references that are in the map. First posted on arXiv 31 Oct 2024.

Builds on

21 of 50 references.

References · 21

  1. 2013Domain Generalization via Invariant Feature Representationcited by 622
  2. 2018Disentangling by Factorisingcited by 614
  3. 2018Understanding disentangling in β-VAEcited by 332
  4. 2017Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributionscited by 125
  5. 2023Nonparametric Identifiability of Causal Representations from Unknown Interventionscited by 105
  6. 2022Linear Causal Disentanglement via Interventionscited by 84
  7. 2021Desiderata for Representation Learning: A Causal Perspectivecited by 76
  8. 2019Causal Discovery with General Non-Linear Relationships Using Non-Linear ICAcited by 58
  9. 2023Score-based Causal Representation Learning with Interventionscited by 49
  10. 2023Causal Discovery with Score Matching on Additive Models with Arbitrary Noisecited by 41
  11. 2021Why did the distribution change?cited by 37
  12. 2023Learning nonparametric latent causal graphs with unknown interventionscited by 34
9 more
  1. 2023General Identifiability and Achievability for Causal Representation Learningcited by 34
  2. 2023Unpaired Multi-Domain Causal Representation Learningcited by 32
  3. 2021Feature Shift Detection: Localizing Which Features Have Shifted via Conditional Distribution Testscited by 24
  4. 2020Causal Mosaic: Cause-Effect Inference via Nonlinear ICA and Ensemble Methodcited by 16
  5. 2018Direct Estimation of Differences in Causal Graphscited by 14
  6. 2023Learning Causal Representations from General Environments: Identifiability and Intrinsic Ambiguitycited by 14
  7. 2023Large Dimensional Independent Component Analysis: Statistical Optimality and Computational Tractabilitycited by 12
  8. 2023Learning DAGs from Data with Few Root Causescited by 8
  9. 2019Direct Learning with Guarantees of the Difference DAG Between Structural Equation Modelscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2025Towards Interpretable Deep Generative Models via Causal Representation Learningcited by 12
  2. 2025Sanity Checking Causal Representation Learning on a Simple Real-World Systemcited by 5
  3. 2026Detecting Changes in Causal Dependence with Kernels and Copulascited by 0
  4. 2026Concept Drift from a Causal Perspectivecited by 0