pith. sign in

arxiv: 2606.06288 · v1 · pith:NMSF5IXLnew · submitted 2026-06-04 · 📊 stat.ML · cs.LG

Discrete Causal Representations from Heterogeneous Domains: A Bayesian Approach with Social Survey Applications

Ankur Garg , Michael Stettler , Aaron Schein , Julius von K\"ugelgen This is my paper

Pith reviewed 2026-06-27 23:19 UTC · model grok-4.3

classification 📊 stat.ML cs.LG
keywords causal representation learningBayesian hierarchical modeldiscrete latent conceptsmulti-environment datasoft interventionssequential Monte Carlosocial survey applications
0
0 comments X

The pith

A Bayesian hierarchical model infers discrete causal concepts and relations from heterogeneous multi-domain data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper builds a hierarchical Bayesian model that turns causal assumptions and interpretability goals into priors and parametric forms for data from multiple environments. It targets the setting of discrete latent concepts with unknown soft interventions that can affect several variables at once. Sequential Monte Carlo sampling approximates the multimodal posterior that results. The method is applied to social survey responses collected across countries or states, where the inferred concepts align with cultural values or political opinions and the relations among them appear plausible. A reader would care because the approach supplies an uncertainty-aware route to high-level causal structure when distribution shifts arise from localized mechanism changes.

Core claim

The paper shows that causal assumptions and interpretability desiderata can be encoded as priors and parametric choices inside a hierarchical model whose multimodal posterior over discrete concepts and unknown multi-node soft interventions is amenable to sequential Monte Carlo approximation, and that the resulting procedure recovers meaningful high-level concepts together with plausible causal relations when applied to multi-country social survey data.

What carries the argument

Hierarchical Bayesian model whose priors encode causal assumptions and interpretability, sampled via sequential Monte Carlo to approximate the posterior over discrete concepts and unknown soft interventions.

If this is right

  • The model identifies discrete high-level causal concepts from low-level measurements even when the data come from different environments.
  • Unknown soft interventions that affect multiple nodes can be handled without prior specification of which nodes are affected.
  • Posterior samples quantify uncertainty over both the concepts and the causal relations among them.
  • Application to social surveys yields concepts that correspond to cultural values or opinions and relations that are consistent with domain knowledge.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same encoding of assumptions into priors could be tested on other heterogeneous data sources such as economic indicators collected across regions.
  • If the SMC approximation proves reliable, the method supplies a practical way to add causal structure to representation learning pipelines that must handle distribution shift.
  • Synthetic benchmarks with known interventions would allow direct measurement of how often the posterior places mass on the true concept graph.

Load-bearing premise

Causal assumptions and interpretability needs can be translated into priors and parametric choices that support valid posterior inference via SMC.

What would settle it

A simulation in which the recovered concepts and relations systematically fail to match the known ground-truth structure used to generate the multi-environment data.

Figures

Figures reproduced from arXiv: 2606.06288 by Aaron Schein, Ankur Garg, Julius von K\"ugelgen, Michael Stettler.

Figure 1
Figure 1. Figure 1: Multi-domain CRL with soft, multi-node interventions. Illustration of our multi￾domain setup with L = 4 causal variables Z = (Z1, Z2, Z3, Z4) and |E| = 3 domains. The intervention targets I e are given by I 1 = {3}, I 2 = {4}, and I 3 = {1, 2}, and the corresponding environment￾specific (changed) mechanisms by p 1 (Z3 | Z1), p 2 (Z4 | Z1, Z2, Z3), and {p 3 (Z1), p3 (Z2)}, respectively. 2.2 Interventional M… view at source ↗
Figure 2
Figure 2. Figure 2: Multi-domain data for CRL with discrete causal latents. We illustrate an example data generating process with L = 3 binary causal latents Z = (Z1, Z2, Z3), a Gaussian measurement model p(X | Z) giving rise to D = 2 continuous observations X = (X1, X2), and |E| = 4 domains (columns; different markers) corresponding to an observational reference setting and interventions on Z1, Z2, and Z3 (from left to right… view at source ↗
Figure 3
Figure 3. Figure 3: Graphical model representation for Bayesian multi-domain CRL with categorical causal latents. Shaded and white circles denote observed and unobserved random variables, respec￾tively. Plates indicate that the contained subgraph and incoming edges are repeated across the corre￾sponding plate indices j, ℓ, e. For each domain e ∈ E, we have a sample of Ne observations x e,j which result from local latent causa… view at source ↗
Figure 4
Figure 4. Figure 4: Illustration of the quadratic KL constraint on δ. (Left, Middle) Samples of δ, colored by the quadratic approximation 1 2 δ ⊤Hδ. The red arrow indicates the constant vector 1 which is orthogonal to the subspace of shift vectors. (Right) Contours of the exact KL divergence and its quadratic approximation, shown on a two-dimensional projection of δ onto its principal components. The (unnormalized) truncated … view at source ↗
Figure 5
Figure 5. Figure 5: Posterior multimodality across permutations of the measurement model. Each red point corresponds to a permutation of the columns of M, with intermediate points obtained by linear interpolations between permutations. The vertical axis shows the log Bayes factor relative to the true ordering. The landscape exhibits multiple local modes, each associated with a distinct causal ordering, while the true ordering… view at source ↗
Figure 6
Figure 6. Figure 6: Decomposition of the log Bayes factors. Each panel shows the contribution of a model component to the evidence difference relative to the true ordering. The left panel shows the measure￾ment model, log p(X, Ψπ | Zπ), which produces multiple posterior modes across permutations. The right panel shows the latent mechanisms, log p(Zπ, Φπ)), which distinguishes among these modes, fa￾voring the true ordering. Th… view at source ↗
Figure 7
Figure 7. Figure 7: Estimated measurement mapping from latent variables to WVS survey responses. Heatmap of the normalized posterior mean absolute measurement matrix, |MT|. The learned block structure is interpretable: Z1 aligns with economic hardship, Z2 with demographics, specifically age and marital status, and Z3 with cultural conservatism. 6.2.1 Interpreting the Measurement Model In this case study, our model associates … view at source ↗
Figure 8
Figure 8. Figure 8: Marginal distributions and baseline causal influence of latent concepts. Left: Country-specific deviations from baseline for each latent concept: EZ∼pe(Z) [Zℓ] − EZ∼p(Z) [Zℓ]. Top Right: Scatterplot of Zhard versus Zcons colored by region. We adopt the regional divisions of the Inglehart–Welzel cultural map (Inglehart et al., 2005), but combine {English-speaking, Protes￾tant Europe, Catholic Europe} into “… view at source ↗
Figure 9
Figure 9. Figure 9: Estimated causal influence of latent concepts across environments (countries). Each panel shows the environment-specific strength of a single dependency. The scale is shared across panels, revealing that influences of Zdemo → Zcons and Zhard → Zcons vary substantially across countries, while the influence of Zhard → Zdemo is relatively small across all environments [PITH_FULL_IMAGE:figures/full_fig_p025_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Intervention indicators and magnitudes across environments. (Top) Posterior mean intervention matrix I, indicating the probability that each latent variable is intervened on in each environment. (Bottom) Expected magnitude of the intervention, as measured by the induced shift in p e (Zℓ), summarizing the sizes of each environment-specific perturbation. 6.4 Interventions In addition to the causal structure… view at source ↗
Figure 11
Figure 11. Figure 11: Intervention directions across parent configurations and environments. Posterior mean intervention vectors δ e cons(pa) for each country, shown for Zdemo = 0 and all configurations of Zhard. Each stem represents an environment; its length indicates intervention magnitude and its orientation indicates direction. Vectors are expressed in an orthogonal basis for the two-dimensional intervention subspace; the… view at source ↗
Figure 12
Figure 12. Figure 12: Posterior summaries for the US political opinion surveys. Each panel shows the learned measurement model (M), intervention probabilities (I), and environment-specific deviations from baseline for each latent variable (EZ∼pe(Z) [Zℓ]−EZ∼p(Z) [Zℓ]). The Trade and Regulation results (a) are largely one-dimensional, separating rural–urban environments along a left–right political gradient. The Elites and Racis… view at source ↗
Figure 13
Figure 13. Figure 13: LLM sampling scheme for generating interventional survey data. We (i) sample demographic covariates from real survey respondents, (ii) use an LLM to expand them into detailed respondent profiles, (iii) apply specified interventions to create multiple intervened versions of each profile, and finally (iv) generate survey responses based on the intervened profiles. This pipeline preserves heterogeneity of th… view at source ↗
Figure 14
Figure 14. Figure 14: “Ground-truth” effects computed from LLM-generated interventional pairs. Each cell shows the average treatment effect for a question–intervention pair, as defined in Eq. (7.1). For example, the bottom–left cells show the mean difference in immigration responses between the Democrat and Republican interventional profiles. The graph is generated by weighting edges according to the average question–level eff… view at source ↗
Figure 15
Figure 15. Figure 15: Estimated measurement mapping for LLM-generated survey responses. Heatmap of the posterior mean absolute measurement matrix, |M|. The learned block structure is interpretable as Z1, Z2, Z3, Z4, Z5 as corresponding to Regulation, Party, Trade, Immigration, and Poverty, respectively, with some Party information also contained in Z1 and Z4. Question top￾ics in this figure and [PITH_FULL_IMAGE:figures/full_f… view at source ↗
Figure 16
Figure 16. Figure 16: Causal graph recovered from LLM responses. Estimated causal influence among la￾tent concepts in the baseline environment, with latent variables labeled according to the block structure identified in [PITH_FULL_IMAGE:figures/full_fig_p032_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Quadratic KL constraint on δ for various θ. We compare the exact KL divergence with second-order approximations based on H0 and Hθ. The H0 approximation deteriorates when θ is far from zero (i.e., highly concentrated distributions), but this regime is unlikely under our prior. B.1 Conjugate Updates B.1.1 immediate conjugacy The following conditional distributions are immediate from standard conjugacy: p(c… view at source ↗
Figure 18
Figure 18. Figure 18: Intervention directions across parent configurations and environments. Pos￾terior mean intervention vectors δ e ℓ (paℓ ) for each country, shown across all parent configurations. Vectors are expressed in an orthogonal basis for the two-dimensional intervention subspace, where the first basis direction captures shifts in the mean of Zℓ and the second captures changes in dispersion. 43 [PITH_FULL_IMAGE:fig… view at source ↗
read the original abstract

Causal representation learning aims to infer the high-level latent causal concepts that give rise to observed low-level measurements. This is particularly relevant for heterogeneous data from different environments or domains since distribution shifts often arise through sparse, localized changes in some of the underlying causal mechanisms, while other parts of the generative process remain unchanged. Whereas identifiability of causal representations has been studied extensively, practical uncertainty-aware methods and real-world use cases remain less explored. In this work, we propose a Bayesian approach to learning causal representations from multi-environment data, focusing on the case of discrete causal concepts and unknown multi-node soft interventions. To this end, we translate causal assumptions and interpretability desiderata into suitable priors and parametric choices within a hierarchical model. We then devise an inference scheme based on sequential Monte Carlo sampling to approximate the resulting multimodal posterior. We showcase our approach through case studies on social survey data, where latent causal concepts correspond to cultural values or political opinions, measurements to survey responses, and environments to different countries or states. Our model infers meaningful high-level concepts and plausible causal relations among them, demonstrating its utility for learning causal representations of complex real-world data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript proposes a Bayesian hierarchical model for causal representation learning from heterogeneous multi-environment data, with a focus on discrete latent causal concepts and unknown multi-node soft interventions. Causal assumptions and interpretability goals are encoded via priors and parametric choices; posterior inference is performed with sequential Monte Carlo (SMC); the approach is demonstrated on social survey data where environments correspond to countries or states, measurements to responses, and latents to cultural or political concepts.

Significance. If the SMC procedure reliably recovers posterior modes that correspond to stable, interpretable causal concepts, the work would supply a practical uncertainty-aware method for causal representation learning on real heterogeneous data, addressing a noted gap between identifiability theory and applied use cases in the social sciences.

major comments (1)
  1. [Inference scheme] Inference section: the central claim that the model infers meaningful high-level concepts and plausible causal relations rests on SMC recovering the modes of a combinatorial, highly multimodal posterior over discrete concepts and unknown multi-node soft interventions. No effective-sample-size diagnostics, tempering schedules, or small-scale exact-enumeration comparisons are reported, leaving open the possibility that reported concepts are sampler artifacts rather than posterior features.
minor comments (1)
  1. [Abstract] Abstract: the description of the modeling choices and inference scheme contains no equations, making it impossible to verify whether the hierarchical construction actually supports the claimed identifiability or posterior properties without the full text.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback. We address the single major comment below.

read point-by-point responses

  1. Referee: [Inference scheme] Inference section: the central claim that the model infers meaningful high-level concepts and plausible causal relations rests on SMC recovering the modes of a combinatorial, highly multimodal posterior over discrete concepts and unknown multi-node soft interventions. No effective-sample-size diagnostics, tempering schedules, or small-scale exact-enumeration comparisons are reported, leaving open the possibility that reported concepts are sampler artifacts rather than posterior features.

    Authors: We agree that the absence of these diagnostics leaves the reliability of the reported modes open to question. In the revised manuscript we will add effective-sample-size diagnostics for the SMC runs on the social-survey data, a description of the tempering schedule, and small-scale synthetic experiments in which the posterior can be enumerated exactly, thereby confirming that the sampler recovers the relevant modes rather than artifacts. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is a standard Bayesian modeling and inference pipeline

full rationale

The paper translates causal assumptions into priors within a hierarchical model and approximates the posterior via SMC. No equations or steps in the abstract reduce a claimed prediction or result to fitted inputs by construction, nor do they rely on self-citations for uniqueness or load-bearing premises. The central claim of inferring concepts is an application of the model to data rather than a tautological renaming or self-definition. The derivation chain is self-contained against external benchmarks of Bayesian causal representation learning.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 1 invented entities

The central claim rests on translating causal assumptions into priors (domain assumption) and on the existence of discrete latent causal concepts as the target of inference (invented entity without independent evidence outside the model). Free parameters are the specific prior hyperparameters chosen to encode interpretability.

free parameters (1)
  • prior hyperparameters
    Chosen by hand to encode causal assumptions and interpretability desiderata in the hierarchical model.
axioms (1)
  • domain assumption Causal assumptions and interpretability desiderata can be translated into suitable priors and parametric choices
    Stated directly in the abstract as the basis for the hierarchical model.
invented entities (1)
  • discrete causal concepts no independent evidence
    purpose: High-level latent variables corresponding to cultural values or political opinions
    Postulated as the target of inference; no independent evidence provided beyond model output on survey data.

pith-pipeline@v0.9.1-grok · 5741 in / 1329 out tokens · 21727 ms · 2026-06-27T23:19:40.835991+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

122 extracted references · 3 canonical work pages

  1. [1]

    Adams, N

    J. Adams, N. Hansen, and K. Zhang. Identification of partially observed linear causal models: Graphical conditions for the non-gaussian and heterogeneous cases. In Advances in Neural Information Processing Systems, volume 34, pages 22822--22833, 2021

    2021

  2. [2]

    Agrawal, C

    R. Agrawal, C. Squires, K. Yang, K. Shanmugam, and C. Uhler. Abcd-strategy: Budgeted experimental design for targeted causal structure discovery. In International Conference on Artificial Intelligence and Statistics, volume 89, pages 3400--3409. PMLR, 2019

    2019

  3. [3]

    Ahuja, J

    K. Ahuja, J. Hartford, and Y. Bengio. Properties from mechanisms: an equivariance perspective on identifiable representation learning. In International Conference on Learning Representations, 2021

    2021

  4. [4]

    Ahuja, D

    K. Ahuja, D. Mahajan, Y. Wang, and Y. Bengio. Interventional causal representation learning. In International conference on machine learning, pages 372--407. PMLR, 2023

    2023

  5. [5]

    Ahuja, A

    K. Ahuja, A. Mansouri, and Y. Wang. Multi-domain causal representation learning via weak distributional invariances. In International Conference on Artificial Intelligence and Statistics, pages 865--873. PMLR, 2024

    2024

  6. [6]

    E. M. Airoldi, D. Blei, E. A. Erosheva, and S. E. Fienberg. Handbook of mixed membership models and their applications. Chapman & Hall/CRC, 1st edition, 2014

    2014

  7. [7]

    J. H. Albert and S. Chib. B ayesian analysis of binary and polychotomous response data. Journal of the American Statistical Association, 88 0 (422): 0 669--679, 1993

    1993

  8. [8]

    J. D. Angrist and J.-S. Pischke. Mostly harmless econometrics: An empiricist's companion . Princeton University Press, 2009

    2009

  9. [9]

    L. P. Argyle, E. C. Busby, N. Fulda, J. R. Gubler, C. Rytting, and D. Wingate. Out of one, many: Using language models to simulate human samples. Political Analysis, 31 0 (3): 0 337--351, 2023

    2023

  10. [10]

    Bareinboim and J

    E. Bareinboim and J. Pearl. Causal inference and the data-fusion problem. Proceedings of the National Academy of Sciences, 113 0 (27): 0 7345--7352, 2016

    2016

  11. [11]

    Bengio, N

    Y. Bengio, N. L \'e onard, and A. Courville. Estimating or propagating gradients through stochastic neurons for conditional computation. arXiv preprint arXiv:1308.3432, 2013

    Pith/arXiv arXiv 2013

  12. [12]

    S. Bing, U. Ninad, J. Wahl, and J. Runge. Identifying linearly-mixed causal representations from multi-node interventions. In Conference on Causal Learning and Reasoning, pages 843--867. PMLR, 2024

    2024

  13. [13]

    D. M. Blei. Build, compute, critique, repeat: Data analysis with latent variable models. Annual Review of Statistics and Its Application, 1 0 (1): 0 203--232, 2014

    2014

  14. [14]

    D. M. Blei, A. Kucukelbir, and J. D. McAuliffe. Variational inference: A review for statisticians. Journal of the American Statistical Association, 112 0 (518): 0 859--877, 2017

    2017

  15. [15]

    Brehmer, P

    J. Brehmer, P. De Haan, P. Lippe, and T. Cohen. Weakly supervised causal representation learning. In Advances in Neural Information Processing Systems, 2022

    2022

  16. [16]

    Brouillard, S

    P. Brouillard, S. Lachapelle, A. Lacoste, S. Lacoste-Julien, and A. Drouin. Differentiable Causal Discovery from Interventional Data . In Advances in Neural Information Processing Systems, volume 33, pages 21865--21877, 2020

    2020

  17. [17]

    Buchholz, G

    S. Buchholz, G. Rajendran, E. Rosenfeld, B. Aragam, B. Sch \"o lkopf, and P. Ravikumar. Learning linear causal representations from interventions under general nonlinear mixing. In Advances in Neural Information Processing Systems, 2023

    2023

  18. [18]

    R. Cai, F. Xie, C. Glymour, Z. Hao, and K. Zhang. Triad constraints for learning causal structure of latent variables. In Advances in Neural Information Processing Systems, volume 32, 2019

    2019

  19. [19]

    C. L. Canonne, I. Diakonikolas, D. M. Kane, and A. Stewart. Testing conditional independence of discrete distributions. In Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing, pages 735--748, 2018

    2018

  20. [20]

    J. Chen, J. Zhu, Z. Wang, X. Zheng, and B. Zhang. Scalable inference for logistic-normal topic models. In C. Burges, L. Bottou, M. Welling, Z. Ghahramani, and K. Weinberger, editors, Advances in Neural Information Processing Systems, volume 26. Curran Associates, Inc., 2013

    2013

  21. [21]

    Chevalley, P

    M. Chevalley, P. Schwab, and A. Mehrjou. Deriving causal order from single-variable interventions: Guarantees & algorithm. In International Conference on Learning Representations, volume 2025, pages 48992--49017, 2025

    2025

  22. [22]

    S. Chib. Marginal likelihood from the gibbs output. Journal of the American Statistical Association, 90 0 (432): 0 1313--1321, 1995

    1995

  23. [23]

    T. M. Cover and J. A. Thomas. Elements of information theory. Wiley-Interscience, Hoboken, N.J, 2nd ed edition, 2006. ISBN 978-0-471-24195-9

    2006

  24. [24]

    C. Dai, J. Heng, P. E. Jacob, and N. Whiteley. An invitation to sequential monte carlo samplers. Journal of the American Statistical Association, 117 0 (539): 0 1587--1600, 2022

    2022

  25. [25]

    H. Dai, Y. Qiu, I. Ng, X. Dong, P. Spirtes, and K. Zhang. Latent variable causal discovery under selection bias. In International Conference on Machine Learning, pages 12161--12178. PMLR, 2025

    2025

  26. [26]

    Daunhawer, A

    I. Daunhawer, A. Bizeul, E. Palumbo, A. Marx, and J. E. Vogt. Identifiability results for multimodal contrastive learning. In The Eleventh International Conference on Learning Representations, 2023

    2023

  27. [27]

    Del Moral, A

    P. Del Moral, A. Doucet, and A. Jasra. Sequential monte carlo samplers. Journal of the Royal Statistical Society Series B: Statistical Methodology, 68 0 (3): 0 411--436, 2006

    2006

  28. [28]

    D. Durante. Conjugate bayes for probit regression via unified skew-normal distributions. Biometrika, 106 0 (4): 0 765--779, 2019

    2019

  29. [29]

    u gelgen, H. Hassani, G. J. Pappas, and B. Sch \

    C. Eastwood, A. Robey, S. Singh, J. Von K \"u gelgen, H. Hassani, G. J. Pappas, and B. Sch \"o lkopf. Probable domain generalization via quantile risk minimization. In Advances in Neural Information Processing Systems, 2022

    2022

  30. [30]

    Eberhardt, C

    F. Eberhardt, C. Glymour, and R. Scheines. On the number of experiments sufficient and in the worst case necessary to identify all causal relations among n variables. In Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence, pages 178--184, 2005

    2005

  31. [31]

    Friedman and D

    N. Friedman and D. Koller. Being B ayesian about network structure. a B ayesian approach to structure discovery in B ayesian networks. Machine learning, 50 0 (1): 0 95--125, 2003

    2003

  32. [32]

    Gelman, J

    A. Gelman, J. B. Carlin, H. S. Stern, D. B. Dunson, A. Vehtari, and D. B. Rubin. B ayesian Data Analysis . Chapman & Hall/CRC Texts in Statistical Science Series. CRC, third edition, 2013. ISBN 9781439840955 1439840954

    2013

  33. [33]

    Gelman, J

    A. Gelman, J. Hwang, and A. Vehtari. Understanding predictive information criteria for bayesian models. Statistics and computing, 24 0 (6): 0 997--1016, 2014

    2014

  34. [34]

    C. J. Geyer. Markov chain monte carlo maximum likelihood. In Computing Science and Statistics: Proceedings of the 23rd Symposium on the Interface, pages 156--163. Interface Foundation of North America, 1991

    1991

  35. [35]

    D. P. Green, B. Palmquist, and E. Schickler. Partisan hearts and minds: Political parties and the social identities of voters. Yale University Press, 2002

    2002

  36. [36]

    Haerpfer, R

    C. Haerpfer, R. Inglehart, A. Moreno, C. Welzel, K. Kizilova, J. Diez-Medrano, M. Lagos, P. Norris, E. Ponarin, B. Puranen, et al. World values survey: Round seven-country-pooled datafile version 5.0, 2022

    2022

  37. [37]

    a gele, J. Rothfuss, L. Lorch, V. R. Somnath, B. Sch \

    A. H \"a gele, J. Rothfuss, L. Lorch, V. R. Somnath, B. Sch \"o lkopf, and A. Krause. Bacadi: B ayesian causal discovery with unknown interventions. In International Conference on Artificial Intelligence and Statistics, pages 1411--1436. PMLR, 2023

    2023

  38. [38]

    Hauser and P

    A. Hauser and P. B \"u hlmann. Characterization and greedy learning of interventional markov equivalence classes of directed acyclic graphs. The Journal of Machine Learning Research, 13 0 (1): 0 2409--2464, 2012

    2012

  39. [39]

    Hauser and P

    A. Hauser and P. B \"u hlmann. Jointly interventional and observational data: estimation of interventional M arkov equivalence classes of directed acyclic graphs. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 77 0 (1): 0 291--318, 2015

    2015

  40. [40]

    Heckerman, D

    D. Heckerman, D. Geiger, and D. M. Chickering. Learning B ayesian networks: The combination of knowledge and statistical data. Machine learning, 20 0 (3): 0 197--243, 1995

    1995

  41. [41]

    Heckerman, C

    D. Heckerman, C. Meek, and G. Cooper. A B ayesian approach to causal discovery. In Innovations in Machine Learning: Theory and Applications, pages 1--28. Springer, 2006

    2006

  42. [42]

    Heinze-Deml, J

    C. Heinze-Deml, J. Peters, and N. Meinshausen. Invariant causal prediction for nonlinear models. Journal of Causal Inference, 6 0 (2), 2018

    2018

  43. [43]

    M. A. Hern\'an and J. M. Robins. Causal inference: What if . Boca Raton: Chapman & Hall/CRC., 2020

    2020

  44. [44]

    Huang, C

    B. Huang, C. J. H. Low, F. Xie, C. Glymour, and K. Zhang. Latent hierarchical causal structure discovery with rank constraints. In Advances in Neural Information Processing Systems, volume 35, pages 5549--5561, 2022

    2022

  45. [45]

    Hyv \"a rinen and H

    A. Hyv \"a rinen and H. Morioka. Unsupervised feature extraction by time-contrastive learning and nonlinear ica. In Advances in Neural Information Processing Systems, pages 3765--3773, 2016

    2016

  46. [46]

    Hyv \"a rinen and H

    A. Hyv \"a rinen and H. Morioka. Nonlinear ICA of temporally dependent stationary sources. In International Conference on Artificial Intelligence and Statistics, pages 460--469. PMLR, 2017

    2017

  47. [47]

    Hyv \"a rinen and P

    A. Hyv \"a rinen and P. Pajunen. Nonlinear independent component analysis: Existence and uniqueness results. Neural networks, 12 0 (3): 0 429--439, 1999

    1999

  48. [48]

    Hyv \"a rinen, H

    A. Hyv \"a rinen, H. Sasaki, and R. Turner. Nonlinear ICA using auxiliary variables and generalized contrastive learning. In International Conference on Artificial Intelligence and Statistics, pages 859--868, 2019

    2019

  49. [49]

    G. W. Imbens and D. B. Rubin. Causal inference in statistics, social, and biomedical sciences . Cambridge University Press, 2015

    2015

  50. [50]

    Inglehart, C

    R. Inglehart, C. Haerpfer, A. Moreno, C. Welzel, K. Kizilova, J. Diez-Medrano, M. Lagos, P. Norris, E. Ponarin, B. Puranen, et al. World values survey, 2005

    2005

  51. [51]

    Jaber, M

    A. Jaber, M. Kocaoglu, K. Shanmugam, and E. Bareinboim. Causal discovery from soft interventions with unknown targets: Characterization and learning. In Advances in Neural Information Processing Systems, volume 33, pages 9551--9561, 2020

    2020

  52. [52]

    Janzing, D

    D. Janzing, D. Balduzzi, M. Grosse-Wentrup, and B. Sch \"o lkopf. Quantifying causal influences. The Annals of Statistics, 41 0 (5): 0 2324--2358, 2013

    2013

  53. [53]

    Jasra, C

    A. Jasra, C. Holmes, and D. Stephens. Markov chain monte carlo methods and the label switching problem in B ayesian mixture modeling. Statistical Science, 20 0 (1): 0 50--67, 2005

    2005

  54. [54]

    Jiang and J

    Z. Jiang and J. Templin. Gibbs samplers for logistic item response models via the p \'o lya--gamma distribution: A computationally efficient data-augmentation strategy. Psychometrika, 84 0 (2): 0 358--374, 2019

    2019

  55. [55]

    Khemakhem, D

    I. Khemakhem, D. Kingma, R. Monti, and A. Hyvarinen. Variational autoencoders and nonlinear ica: A unifying framework. In International Conference on Artificial Intelligence and Statistics, pages 2207--2217, 2020

    2020

  56. [56]

    D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. In International Conference on Learning Representations (ICLR), 2015. URL https://arxiv.org/abs/1412.6980

    Pith/arXiv arXiv 2015

  57. [57]

    Kitagawa

    G. Kitagawa. Monte carlo filter and smoother for non-gaussian nonlinear state space models. Journal of Computational and Graphical Statistics, 5 0 (1): 0 1--25, 1996

    1996

  58. [58]

    Kivva, G

    B. Kivva, G. Rajendran, P. Ravikumar, and B. Aragam. Learning latent causal graphs via mixture oracles. In Advances in Neural Information Processing Systems, volume 34, pages 18087--18101, 2021

    2021

  59. [59]

    Krueger, E

    D. Krueger, E. Caballero, J.-H. Jacobsen, A. Zhang, J. Binas, D. Zhang, R. Le Priol, and A. Courville. Out-of-distribution generalization via risk extrapolation (rex). In International Conference on Machine Learning, pages 5815--5826. PMLR, 2021

    2021

  60. [60]

    Lachapelle, P

    S. Lachapelle, P. R. L \'o pez, Y. Sharma, K. Everett, R. L. Priol, A. Lacoste, and S. Lacoste-Julien. Nonparametric partial disentanglement via mechanism sparsity: Sparse actions, interventions and sparse temporal dependencies. arXiv preprint arXiv:2401.04890, 2024

    arXiv 2024

  61. [61]

    and Romano, Joseph P

    E. Lehmann and J. P. Romano. Testing Statistical Hypotheses. Springer Texts in Statistics. Springer International Publishing, Cham, 2022. ISBN 978-3-030-70577-0. doi:10.1007/978-3-030-70578-7. URL https://link.springer.com/10.1007/978-3-030-70578-7

    work page doi:10.1007/978-3-030-70578-7 2022

  62. [62]

    J. Li, R. Gibbons, and V. Ro c kov \'a . Sparse B ayesian multidimensional item response theory. Journal of the American Statistical Association, 120 0 (552): 0 2592--2605, 2025

    2025

  63. [63]

    Lippe, S

    P. Lippe, S. Magliacane, S. L \"o we, Y. M. Asano, T. Cohen, and E. Gavves. Causal representation learning for instantaneous and temporal effects in interactive systems. In International Conference on Learning Representations, 2022 a

    2022

  64. [64]

    Lippe, S

    P. Lippe, S. Magliacane, S. L \"o we, Y. M. Asano, T. Cohen, and S. Gavves. Citris: Causal identifiability from temporal intervened sequences. In International Conference on Machine Learning, pages 13557--13603. PMLR, 2022 b

    2022

  65. [65]

    Locatello, S

    F. Locatello, S. Bauer, M. Lucic, G. Raetsch, S. Gelly, B. Sch \"o lkopf, and O. Bachem. Challenging common assumptions in the unsupervised learning of disentangled representations. In International Conference on Machine Learning, pages 4114--4124. PMLR, 2019

    2019

  66. [66]

    Lorch, J

    L. Lorch, J. Rothfuss, B. Sch \"o lkopf, and A. Krause. DiBS : Differentiable B ayesian structure learning. In Advances in Neural Information Processing Systems, volume 34, pages 24111--24123, 2021

    2021

  67. [67]

    F. M. Lord. Applications of item response theory to practical testing problems. Routledge, 1980

    1980

  68. [68]

    C. J. Maddison, A. Mnih, and Y. W. Teh. The concrete distribution: A continuous relaxation of discrete random variables. arXiv preprint arXiv:1611.00712, 2016

    Pith/arXiv arXiv 2016

  69. [69]

    Marinari and G

    E. Marinari and G. Parisi. Simulated tempering: a new monte carlo scheme. EPL (Europhysics Letters), 19 0 (6): 0 451--458, 1992

    1992

  70. [70]

    Mason and J

    L. Mason and J. Wronski. One tribe to bind them all: How our social group attachments strengthen partisanship. Political Psychology, 39: 0 257--277, 2018

    2018

  71. [71]

    Moran and B

    G. Moran and B. Aragam. Towards interpretable deep generative models via causal representation learning. Journal of the American Statistical Association, pages 1--32, 2026

    2026

  72. [72]

    S. L. Morgan and C. Winship. Counterfactuals and Causal Inference: Methods and Principles for Social Research . Cambridge University Press, 2014

    2014

  73. [73]

    K. P. Murphy. Active learning of causal B ayes net structure. Technical report, Department of Computer Science, U.C. Berkeley, 2001

    2001

  74. [74]

    R. M. Neal. Sampling from multimodal distributions using tempered transitions. Statistics and computing, 6 0 (4): 0 353--366, 1996

    1996

  75. [75]

    R. O. Ness, K. Sachs, P. Mallick, and O. Vitek. A bayesian active learning experimental design for inferring signaling networks. Journal of Computational Biology, 25 0 (7): 0 709--725, 2018

    2018

  76. [76]

    Neykov, S

    M. Neykov, S. Balakrishnan, and L. Wasserman. Minimax optimal conditional independence testing. The Annals of Statistics, 49 0 (4): 0 2151--2177, 2021

    2021

  77. [77]

    J. Neyman. Sur les applications de la th \'e orie des probabilit \'e s aux experiences agricoles: Essai des principes. Roczniki Nauk Rolniczych, 10 0 (1): 0 1--51, 1923

    1923

  78. [78]

    J. Pearl. Causality: Models, Reasoning, and Inference. Cambridge University Press, New York, NY, 2nd edition, 2009

    2009

  79. [79]

    u gelgen, and B. Sch \

    R. Perry, J. Von K \"u gelgen, and B. Sch \"o lkopf. Causal discovery in heterogeneous environments under the sparse mechanism shift hypothesis. In Advances in Neural Information Processing Systems, volume 35, pages 10904--10917, 2022

    2022

  80. [80]

    Peters, P

    J. Peters, P. B \"u hlmann, and N. Meinshausen. Causal inference by using invariant prediction: identification and confidence intervals. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 78 0 (5): 0 947--1012, 2016

    2016

Showing first 80 references.