The reviewed record of science sign in
Pith

arxiv: 2606.28854 · v1 · pith:OFBW2EXF · submitted 2026-06-27 · stat.ML · cs.AI· cs.LG· math.ST· stat.CO· stat.TH

Perspectives on Latent Factor Indeterminacy and its Implications for Data Representation

Carel F.W. Peeters This is my paper

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel 2026-06-30 08:52 UTCgrok-4.3pith:OFBW2EXFrecord.jsonopen to challenge →

classification stat.ML cs.AIcs.LGmath.STstat.COstat.TH
keywords latent factor indeterminacyfactor analysisdata representationgenerative neural networksautoencodershigh-dimensional datavariational autoencoderslatent variable collapse
0
0 comments X

The pith

Latent factor indeterminacy resolves across all facets once the number of features grows to infinity.

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

The paper establishes that indeterminacy in retrieving unique latent sources from factor models disappears when the observed feature dimension grows without bound. This holds even after accounting for rotational indeterminacy and known intrinsic dimension, yielding an essentially distribution-free estimation method in large-feature samples. A sympathetic reader would care because the result positions the factor model as a viable basal learner for representing very high-dimensional data and links classical psychometric indeterminacy to modern latent collapse problems in variational autoencoders.

Core claim

The common factor analytic model can be viewed as a linear autoencoder or single-hidden-layer generative neural network. Under this view, the paper shows that latent factor determinacy holds across all facets when feature dimension tends to infinity. This supplies a distribution-free estimation route in the sample case for very large feature counts and implies the factor model is suited for representation learning of very-high-dimensional data.

What carries the argument

The common factor analytic model conceived as a linear autoencoder or single-hidden-layer generative neural network, which transfers classical indeterminacy results into modern generative frameworks.

If this is right

  • An essentially distribution-free estimation approach becomes available when the number of features is very large.
  • The factor model is suited for representation learning of very-high-dimensional data as an emergent property at scale.
  • Indeterminacy issues that appear in related frameworks such as variational autoencoders are expected to diminish at large feature counts.
  • The result carries implications for data representation practices in psychometrics, statistics, and artificial intelligence.

Where Pith is reading between the lines

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

  • High-dimensional datasets from genomics or imaging could serve as natural test beds to verify whether the distribution-free estimator recovers stable latents in practice.
  • The same scaling argument may extend to deeper generative architectures if they contain a factor-analytic bottleneck layer.
  • Practitioners working with moderate feature counts might still need explicit regularization or side constraints to mitigate remaining indeterminacy.

Load-bearing premise

The common factor analytic model can be conceived as a linear autoencoder or single-hidden-layer generative neural network so that classical indeterminacy results transfer directly.

What would settle it

A concrete counter-example in which indeterminacy persists, or the distribution-free estimator fails to recover consistent latent sources, even as the number of features tends to infinity.

Figures

Figures reproduced from arXiv: 2606.28854 by Carel F.W. Peeters.

Figure 1
Figure 1. Figure 1: Schematic of the common factor model. The nodes indicate our random variables, with [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: A visualization of the correlational geometry of indeterminacy. The angles [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Visualization of the Bayesian view on indeterminacy in one dimension involving [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
read the original abstract

The common factor analytic model is related to Helmholtz and Boltzmann machines, can be conceived as a linear autoencoder, or can be thought of as a single-hidden-layer generative neural network. We thus consider it a basal generative representation learner that can be used as a minimal model for studying the foundational characteristics of (deep) generative model architectures. We focus on the fundamental problem of indeterminacy in latent factor projections. This indeterminacy implies that, even when the intrinsic dimension of the latent vector is known, regularity conditions are met, and rotational indeterminacy is resolved, an inherent indefiniteness in the retrieval of causative latent sources remains: they will be uncertain, distributionally deviant, and non-unique. This can have major implications for data representation but remains an elusive issue, even to practitioners and theorists well-versed in the factor model. Moreover, this classic psychometric problem is intricately related to the modern issue of latent variable collapse in the variational autoencoder framework for deep generative modeling. Here, we assess this indeterminacy from various perspectives and show how these are mathematically and conceptually related and we discuss subsequent implications for the Psychometrics, Statistics, and Artificial Intelligence communities. We show that one has latent factor determinacy across all its facets when the feature-dimension grows to infinity. This feeds into an essentially distribution-free estimation approach in the sample case when the number of features grows very large. We conclude, as these are emergent properties at scale, that the factor model is suited for representation learning of very-high-dimensional 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

2 major / 1 minor

Summary. The manuscript relates the common factor analytic model to Helmholtz/Boltzmann machines, linear autoencoders, and single-hidden-layer generative networks, treating it as a basal model for studying indeterminacy in latent factor projections. It examines multiple perspectives on this indeterminacy (uncertainty, distributional deviation, non-uniqueness), links it to VAE latent collapse, and claims that determinacy across all facets emerges when the feature dimension p tends to infinity, yielding an essentially distribution-free estimation approach for representation learning of very-high-dimensional data.

Significance. If the determinacy result at infinite dimension is rigorously established and the transfer from the linear single-layer case to modern deep generative models is justified, the work could provide a foundational bridge between classical psychometrics and high-dimensional representation learning in AI, with the distribution-free estimation at scale as a potentially useful emergent property.

major comments (2)
  1. [Abstract] Abstract: The central claim that 'one has latent factor determinacy across all its facets when the feature-dimension grows to infinity' and that this 'feeds into an essentially distribution-free estimation approach' is load-bearing for the paper's conclusions, yet the provided text contains no equations, derivations, or proofs to verify the p→∞ limit or its resolution of non-uniqueness and distributional deviation.
  2. [Abstract] Abstract: The premise that the factor model conceived as a linear autoencoder or single-hidden-layer generative NN allows classical indeterminacy results to transfer directly to deep generative frameworks (including VAE collapse) is load-bearing for the implications to AI and representation learning; however, this equivalence is strictly linear and single-layer, while deep models involve non-linear activations and stacked layers whose indeterminacy properties are not automatically resolved by the same p→∞ limit.
minor comments (1)
  1. [Abstract] The abstract uses several near-synonyms for indeterminacy ('uncertain, distributionally deviant, and non-unique') without defining how these facets are mathematically distinguished or related.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their detailed and constructive report. We address each major comment point by point below, clarifying the scope and content of the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that 'one has latent factor determinacy across all its facets when the feature-dimension grows to infinity' and that this 'feeds into an essentially distribution-free estimation approach' is load-bearing for the paper's conclusions, yet the provided text contains no equations, derivations, or proofs to verify the p→∞ limit or its resolution of non-uniqueness and distributional deviation.

    Authors: The abstract is a high-level summary and does not contain equations by design. The full manuscript derives the p→∞ determinacy result, showing that uncertainty, distributional deviation, and non-uniqueness all resolve in this limit, and connects this to the distribution-free estimation property for large feature dimensions. We can revise the abstract to include a brief pointer to the relevant theorem if the editor prefers. revision: partial

  2. Referee: [Abstract] Abstract: The premise that the factor model conceived as a linear autoencoder or single-hidden-layer generative NN allows classical indeterminacy results to transfer directly to deep generative frameworks (including VAE collapse) is load-bearing for the implications to AI and representation learning; however, this equivalence is strictly linear and single-layer, while deep models involve non-linear activations and stacked layers whose indeterminacy properties are not automatically resolved by the same p→∞ limit.

    Authors: The manuscript positions the factor model as a basal generative representation learner for studying foundational indeterminacy properties, not as a direct equivalent to non-linear deep networks. The p→∞ result is established in the linear case, with implications for high-dimensional representation learning discussed as emergent properties at scale. We do not claim automatic resolution of indeterminacy in arbitrary stacked non-linear models. revision: no

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained against external benchmarks

full rationale

The paper premises the factor model as equivalent to a linear autoencoder or single-hidden-layer generative NN to study indeterminacy, then states a determinacy result as p→∞ that feeds into distribution-free estimation. No equations or steps are quoted that reduce the claimed determinacy (or its transfer) to a fitted parameter, self-defined quantity, or self-citation chain by construction. The relation to VAEs is conceptual framing rather than a load-bearing mathematical reduction. The result is presented as shown from the high-dimensional limit, independent of the paper's own fitted values or prior self-citations in the visible text. This meets the criteria for a self-contained derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only view yields no identifiable free parameters, axioms, or invented entities; ledger left empty due to insufficient technical detail.

pith-pipeline@v0.9.1-grok · 5811 in / 918 out tokens · 33320 ms · 2026-06-30T08:52:49.487907+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

164 extracted references

  1. [1]

    barticle [author] Ackley , D. H. D. H. , Hinton , G. E. G. E. Sejnowksi , T. J. T. J. ( 1985 ). A learning algorithm for Boltzmann machines . Cognitive Science 9 147--169 . barticle

    1985

  2. [2]

    bincollection [author] Aitkin , M. M. Aitkin , I. I. ( 2005 ). Bayesian inference for factor scores . In Contemporary Psychometrics: A Festschrift for Roderick P.\ McDonald ( A. A. Maydeu-Olivares J. J. J. J. McArdle , eds.) 207--221 . Mahwah New Jersey: Lawrence Erlbaum Associates . bincollection

    2005

  3. [3]

    binproceedings [author] Alemi , A. A. A. A. , Poole , B. B. , Fischer , I. I. , Dillon , J. V. J. V. , Saurous , R. A. R. A. Murphy , K. K. ( 2018 ). Fixing a broken ELBO In Proceedings of the 35th International Conference on Machine Learning . binproceedings

    2018

  4. [4]

    barticle [author] Amari , S. S. ( 1967 ). A theory of adaptive patterns classifier . IEEE Transactions on Electronic Computers EC-16 299--307 . barticle

    1967

  5. [5]

    bbook [author] Anderson , T. W. T. W. ( 2003 ). An Introduction to Multivariate Statistical Analysis . Hoboken, NJ : J ohn W iley & S ons, I nc. (3rd ed.) . bbook

    2003

  6. [6]

    binproceedings [author] Anderson , T. W. T. W. Rubin , H. H. ( 1956 ). Statistical Inference in Factor Analysis . In Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability 5: Contributions to Econometrics, Industrial Research, and Psychometry 111--150 . Berkeley, CA: University of California Press . binproceedings

    1956

  7. [7]

    barticle [author] Baglama , J. J. Reichel , L. L. ( 2005 ). Augmented implicitly restarted Lanczos bidiagonalization methods . SIAM Journal on Scientific Computing 27 19--42 . barticle

    2005

  8. [8]

    barticle [author] Bartholomew , D. J. D. J. ( 1981 ). Posterior analysis of the factor model . British Journal of Mathematical and Statistical Psychology 34 93-99 . barticle

    1981

  9. [9]

    bbook [author] Bekker , P. A. P. A. , Merckens , A. A. Wansbeek , T. J. T. J. ( 1994 ). Identification, Equivalent Models, and Computer Algebra . Boston [etc.]: A cademic P ress . bbook

    1994

  10. [10]

    bbook [author] Bishop , C. M. C. M. Bishop , H. H. ( 2024 ). Deep Learning: Foundations and Concepts . Springer . bbook

    2024

  11. [11]

    bbook [author] Bollen , K. A. K. A. ( 1989 ). Structural Equations with Latent Variables . N ew Y ork [etc.]: J ohn W iley & S ons . bbook

    1989

  12. [12]

    barticle [author] Borsboom , D. D. , Mellenbergh , G. J. G. J. van Heerden , J. J. ( 2003 ). The theoretical status of latent variables . Psychological Review 110 203--219 . barticle

    2003

  13. [13]

    binproceedings [author] Bowman , S. R. S. R. , Vilnis , L. L. , Vinyals , O. O. , Dai , A. M. A. M. , Jozefowicz , R. R. Bengio , S. S. ( 2016 ). Generating sentences from a continuous space In SIGNLL Conference on Computational Natural Language Learning (CONLL), 2016 . binproceedings

    2016

  14. [14]

    bbook [author] Cattell , R. B. R. B. ( 1978 ). The Scientific Use of Factor Analysis in Behavioral and Life Sciences . New York: Plenum Press . bbook

    1978

  15. [15]

    barticle [author] Chen , B. B. , Polatkan , G. G. , Shapiro , G. G. , Blei , D. D. , Dunson , D. D. Carin , L. L. ( 2013 ). Deep learning with hierarchical convolutional factor analysis . IEEE Transactions on Pattern Analysis and Machine Intelligence 35 1887--1901 . barticle

    2013

  16. [16]

    bbook [author] Cram\' e r , H. H. ( 1946 ). Mathematical Methods of Statistics . Princeton: Princeton University Press . bbook

    1946

  17. [17]

    bincollection [author] Cueto , M. A. M. A. , Morton , J. J. Sturmfels , B. B. ( 2010 ). Geometry of the restricted Boltzmann machine . In Algebraic Methods in Statistics and Probability II , ( M. M. Viana H. H. Wynn , eds.). Contemporary Mathematics 516 135--153 . Providence: American Mathematical Society . bincollection

    2010

  18. [18]

    binproceedings [author] Dai , B. B. , Wang , Z. Z. Wipf , D. D. ( 2020 ). The usual suspects? Reassessing blame for VAE posterior collapse In Proceedings of the 37th International Conference on Machine Learning . binproceedings

    2020

  19. [19]

    barticle [author] Dai , F. F. , Dutta , S. S. Maitra , R. R. ( 2020 ). A matrix-free likelihood method for exploratory factor analysis of high-dimensional G aussian data . Journal of Computational and Graphical Statistics 29 675--680 . barticle

    2020

  20. [20]

    barticle [author] Dayan , P. P. Hinton , G. E. G. E. ( 1996 ). Varieties of Helmholtz machine . Neural Networks 8 1385--1403 . barticle

    1996

  21. [21]

    barticle [author] del Pino , G. E. G. E. Galaz , H. H. ( 1995 ). Statistical applications of the inverse Gram matrix: A revisitation . Brazilian Journal of Probability and Statistics 9 177--196 . barticle

    1995

  22. [22]

    binproceedings [author] Delalleau , O. O. Bengio , Y. Y. ( 2011 ). Shallow vs. deep sum-product networks In Advances in Neural Information Processing Systems 24 (NIPS 2011) . binproceedings

    2011

  23. [23]

    barticle [author] Diaconis , P. P. Freedman , D. D. ( 1984 ). Asymptotics of graphical projection pursuit . Annals of Statistics 12 793--815 . barticle

    1984

  24. [24]

    bbook [author] Dieudonn\' e , J. J. ( 1963 ). La G\' e om\' e trie des Groupes Classiques . Berlin: Springer-Verlag (2nd ed.) . bbook

    1963

  25. [25]

    bmisc [author] Donoho , D. L. D. L. ( 2000 ). High-dimensional data analysis: The curses and blessings of dimensionality. Invited lecture at Mathematical Challenges of the 21st Century, AMS National Meeting, Los Angeles, CA, USA . bmisc

    2000

  26. [26]

    barticle [author] Duncan , W. J. W. J. ( 1944 ). Some devices for the solution of large sets of simultaneous linear equations (with an appendix on the reciprocation of partitioned matrices) . London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science: Series 7 35 660--670 . barticle

    1944

  27. [27]

    barticle [author] Elffers , H. H. , Bethlehem , J. J. Gill , R. R. ( 1978 ). Indeterminacy problems and the interpretation of factor analysis results . Statistica Neerlandica 32 181--199 . barticle

    1978

  28. [28]

    barticle [author] Erhan , D. D. , Bengio , Y. Y. , Courville , A. A. , Manzagol , P. A. P. A. , Vincent , P. P. Bengio , S. S. ( 2010 ). Why does unsupervised pre-training help deep learning? Journal of Machine Learning Research 11 625--660 . barticle

    2010

  29. [29]

    barticle [author] Fefferman , C. C. , Mitter , S. S. Narayanan , H. H. ( 2016 ). Testing the manifold hypothesis . Journal of the American Mathematical Society 29 983--1049 . barticle

    2016

  30. [30]

    barticle [author] Geweke , J. F. J. F. Singleton , K. J. K. J. ( 1980 ). Interpreting the likelihood ratio statistic in factor models when sample size is small . Journal of the American Statistical Association 75 133--137 . barticle

    1980

  31. [31]

    bbook [author] Goodfellow , I. I. , Bengio , Y. Y. Courville , A. A. ( 2016 ). Deep Learning . Cambridge, MT: MIT Press . bbook

    2016

  32. [32]

    barticle [author] Gorban , A. N. A. N. Tyukin , I. Y. I. Y. ( 2018 ). Blessing of dimensionality: Mathematical foundations of the statistical physics of data . Philosophical Transactions of the Royal Society A 376 20170237 . barticle

    2018

  33. [33]

    bbook [author] Gut , A. A. ( 2005 ). Probability: A Graduate Course . New York: Springer . bbook

    2005

  34. [34]

    barticle [author] Guttman , L. L. ( 1955 ). The determinacy of factor score matrices with implications for five other basic problems of common factor theory . British Journal of Mathematical and Statistical Psychology 8 65--81 . barticle

    1955

  35. [35]

    barticle [author] Haig , B. D. B. D. ( 2005 ). Exploratory factor analysis, theory generation, and scientific method . Multivariate Behavioral Research 40 303--329 . barticle

    2005

  36. [36]

    barticle [author] Hall , P. P. Li , K. C. K. C. ( 1993 ). On almost linearity of low-dimensional projections from high-dimensional data . Annals of Statistics 21 867--889 . barticle

    1993

  37. [37]

    barticle [author] Henderson , H. V. H. V. Searle , S. R. S. R. ( 1981 ). On deriving the inverse of a sum of matrices . SIAM Review 23 53--60 . barticle

    1981

  38. [38]

    barticle [author] Hendrickson , A. E. A. E. White , P. O. P. O. ( 1964 ). PROMAX: A quick method for rotation to oblique simple structure . British Journal of Mathematical & Statistical Psychology 17 65--70 . barticle

    1964

  39. [39]

    barticle [author] Hinton , G. E. G. E. Ghahramani , Z. Z. ( 1997 ). Generative models for discovering sparse distributed representations . Philosophical Transactions of the Royal Society of London, B 352 1177--1190 . barticle

    1997

  40. [40]

    barticle [author] Hinton , G. E. G. E. Salakhutdinov , R. R. R. R. ( 2006 ). Reducing the dimensionality of data with neural networks . Science 313 504--507 . barticle

    2006

  41. [41]

    barticle [author] Hirose , K. K. Yamamoto , M. M. ( 2015 ). Sparse estimation via nonconcave penalized likelihood in factor analysis model . Statistics & Computing 25 863--875 . barticle

    2015

  42. [42]

    bbook [author] Horn , R. A. R. A. Johnson , C. A. C. A. ( 1985 ). Matrix Analysis . Cambridge: Cambdridge University Press . bbook

    1985

  43. [43]

    barticle [author] Hornik , K. K. , Stinchcombe , M. M. White , H. H. ( 1989 ). Multilayer feedforward networks are universal approximators . Neural Networks 2 359--366 . barticle

    1989

  44. [44]

    btechreport [author] Howe , W. G. W. G. ( 1955 ). Some contributions to factor analysis. Technical Report No. ORNL-1919 , Oak Ridge National Laboratory . btechreport

    1955

  45. [45]

    bbook [author] Ivakhnenko , A. G. A. G. Lapa , V. G. V. G. ( 1967 ). Cybernetics and Forecasting Techniques . New York: American Elsevier Publishing co. bbook

    1967

  46. [46]

    barticle [author] Jennrich , R. I. R. I. Robinson , S. M. S. M. ( 1969 ). A Newton-Rhapson algorithm for maximum likelihood factor analysis . Psychometrika 34 111--123 . barticle

    1969

  47. [47]

    barticle [author] J\" o reskog , K. G. K. G. ( 1967 ). Some contributions to maximum likelihood factor analysis . Psychometrika 32 443--482 . barticle

    1967

  48. [48]

    bincollection [author] Kainen , P. C. P. C. ( 1997 ). Utilizing Geometric Anomalies of High Dimension: When Complexity Makes Computation Easier . In Computer Intensive Methods in Control and Signal Processing ( M. M. K\' a rn\' y K. K. Warwick , eds.) 18 , 283--294 . Boston, MA: Birkh\" a user . bincollection

    1997

  49. [49]

    barticle [author] Kainen , P. C. P. C. K u rkov\' a , V. V. ( 1993 ). Quasiorthogonal dimension of Euclidian spaces . Applied Mathematics Letters 6 7--10 . barticle

    1993

  50. [50]

    barticle [author] Kaiser , H. F. H. F. ( 1958 ). The Varimax criterion for analytic rotation in factor analysis . Psychometrika 23 187--200 . barticle

    1958

  51. [51]

    barticle [author] Kaiser , H. F. H. F. Rice , J. J. ( 1974 ). Little jiffy, mark IV . Educational & Psychological Measurement 34 111--117 . barticle

    1974

  52. [52]

    bbook [author] Kato , T. T. ( 1976 ). Perturbation Theory for Linear Operators . New-York: Springer-Verlag . bbook

    1976

  53. [53]

    barticle [author] Kestelman , H. H. ( 1952 ). The fundamental equation of factor analysis . British Journal of Mathematical and Statistical Psychology 5 1--6 . barticle

    1952

  54. [54]

    binproceedings [author] Kingma , D. D. , Salimans , T. T. , Josefowicz , R. R. , Chen , X. X. , Sutskever , I. I. Welling , M M. ( 2016 ). Improving variational autoencoders with inverse autoregressive flow . In 30th Annual Conference on Neural Information Processing Systems ( D. D. Lee , U U. von Luxburg , R. R. Garnett , M. M. Sugiyama I. I. Guyon , eds...

    2016

  55. [55]

    binproceedings [author] Kingma , D. P. D. P. Welling , M. M. ( 2014 ). Auto-encoding variational Bayes In International Conference on Learning Representations (ICLR) 2014 . binproceedings

    2014

  56. [56]

    barticle [author] Kingma , D. P. D. P. Welling , M. M. ( 2019 ). An introduction to variational autoencoders . Foundations and Trends in Machine Learning 12 307--392 . barticle

    2019

  57. [57]

    barticle [author] Kong , X. X. , Jiang , X. X. , Zhang , B. B. , Yuan , J. J. Ge , Z. Z. ( 2022 ). Latent variable models in the era of industrial big data: Extension and beyond . Annual Reviews in Control 54 167--199 . barticle

    2022

  58. [58]

    barticle [author] Koster , J. T. A. J. T. A. ( 1999 ). On the validity of the Markov interpretation of path diagrams of Gaussian structural equations systems with correlated errors . Scandinavian Journal of Statistics 26 413--431 . barticle

    1999

  59. [59]

    barticle [author] Krijnen , W. P. W. P. , Dijkstra , T. K. T. K. Gill , R. D. R. D. ( 1998 ). Conditions for factor (in)determinacy in factor analysis . Psychometrika 63 359--367 . barticle

    1998

  60. [60]

    barticle [author] Kuo , Z. Y. Z. Y. ( 1921 ). Giving up instincts in psychology . Journal of Philosophy 18 645--664 . barticle

    1921

  61. [61]

    barticle [author] Lawley , D. N. D. N. ( 1940 ). The estimation of factor loadings by the method of Maximum Likelihood . Proceedings of the Royal Society of Edinburgh 60 64--82 . barticle

    1940

  62. [62]

    binproceedings [author] Lucas , J. J. , Tucker , G. G. , Grosse , R. R. Narauzi , M. M. ( 2019 ). Understanding posterior collapse in generative latent variable models In International Conference on Learning Representations (ICLR) 2019 . binproceedings

    2019

  63. [63]

    barticle [author] Maraun , M. D. M. D. ( 1996 ). Metaphor taken as math: I ndeterminancy in the factor analysis model . Multivariate Behavioral Research 31 517--538 . barticle

    1996

  64. [64]

    barticle [author] McCrimmon , K. K. ( 1978 ). Jordan algebras and their applications . Bulletin of the American Mathematical Society 84 612--627 . barticle

    1978

  65. [65]

    barticle [author] McDonald , R. P. R. P. Burr , E. J. E. J. ( 1967 ). A comparison of four methods of constructing factor scores . Psychometrika 32 381--401 . barticle

    1967

  66. [66]

    barticle [author] McDonald , R. P. R. P. Mulaik , S. A. S. A. ( 1979 ). Determinacy of common factors: A nontechnical review . Psychological Bulletin 86 297--306 . barticle

    1979

  67. [67]

    barticle [author] McNeish , D. D. Wolf , M. G. M. G. ( 2020 ). Thinking twice about sum scores . Behavior Research Methods 52 2287--2305 . barticle

    2020

  68. [68]

    bbook [author] Milman , V. D. V. D. Schechtman , G. G. ( 1986 ). Asymptotic Theory of Finite Dimensional Normed Spaces . Lecture Notes in Mathematics 1200 . Berlin: Springer-Verlag . bbook

    1986

  69. [69]

    barticle [author] Mulaik , S. A. S. A. ( 1976 ). Comments on ``The measurement of factorial indeterminacy" . Psychometrika 41 249--262 . barticle

    1976

  70. [70]

    bincollection [author] Mulaik , S. A. S. A. ( 2005 ). Looking back on the indeterminacy controversies in factor analysis . In Contemporary Psychometrics: A Festschrift for Roderick P.\ McDonald ( A. A. Maydeu-Olivares J. J. J. J. McArdle , eds.) 7 , 174--206 . Mahwah New Jersey: Lawrence Erlbaum Associates . bincollection

    2005

  71. [71]

    bbook [author] Mulaik , S. A. S. A. ( 2010 ). Foundations of Factor Analysis . Boca Raton, FL: Chapman Hall/CRC (2nd ed.) . bbook

    2010

  72. [72]

    barticle [author] M\" u nch , M. M. M. M. , van de Wiel , M. A. M. A. , van der Vaart , A. W. A. W. Peeters , C. F. W. C. F. W. ( 2022 ). Semi-supervised empirical Bayes group-regularized factor regression . Biometrical Journal 64 1289--1306 . barticle

    2022

  73. [73]

    barticle [author] Nakajima , S. S. , Tomioka , R. R. , Sugiyama , M. M. Babacan , S. D. S. D. ( 2015 ). Condition for perfect dimensionality recovery by variational Bayesian PCA . Journal of Machine Learning Research 16 3757--3811 . barticle

    2015

  74. [74]

    bbook [author] Pearl , J. J. ( 2009 ). Causality: Models, Reasoning, and Inference . Cambridge: Cambridge University Press (2nd ed.) . bbook

    2009

  75. [75]

    bphdthesis [author] Peeters , C. F. W. C. F. W. ( 2012 ). Bayesian Exploratory and Confirmatory Factor Analysis: Perspectives on Constrained-Model Selection , PhD thesis , Dept.\ of Methodology & Statistics, Utrecht University . bphdthesis

    2012

  76. [76]

    barticle [author] Peeters , C. F. W. C. F. W. ( 2012 ). Rotational uniqueness conditions under oblique factor correlation metric . Psychometrika 77 288--292 . barticle

    2012

  77. [77]

    barticle [author] Piaggio , H. T. H. H. T. H. ( 1931 ). The general factor in Spearman's theory of intelligence . Nature 127 56--57 . barticle

    1931

  78. [78]

    barticle [author] Piaggio , H. T. H. H. T. H. ( 1933 ). Three sets of conditions necessary for the existence of a g that is real and unique except in sign . British Journal of Psychology 24 88--105 . barticle

    1933

  79. [79]

    barticle [author] Rao , C. R. C. R. ( 1955 ). Estimation and tests of significance in factor analysis . Psychometrika 20 93--111 . barticle

    1955

  80. [80]

    barticle [author] Rao , R. P. R. P. Bollard , D. H. D. H. ( 1999 ). Predictive coding in the visual cortex: A functional interpretation of some extra-classical receptive field effects . Nature Neuroscience 2 79--87 . barticle

    1999

Showing first 80 references.