{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T09:51:19Z","timestamp":1791280279889,"version":"4.1.0"},"reference-count":59,"publisher":"Cambridge University Press (CUP)","issue":"4","license":[{"start":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T00:00:00Z","timestamp":1778112000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["cambridge.org"],"crossmark-restriction":true},"short-container-title":["Psychometrika"],"published-print":{"date-parts":[[2026,9]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Recent research shows that amortized variational inference (AVI) can be used to efficiently estimate high-dimensional latent variable models on large datasets. However, its use has remained limited to item response theory (IRT), and generalizing the approach to discrete latent variable models is not straightforward. We propose two ways to deal with this problem. In an initial simulation, we verify that these approaches can be used to estimate simple discrete latent variable models, such as latent class analysis and the generalized deterministic inputs, noisy and gate model. In these cases, AVI provides accurate parameter estimates, although the computational advantage over marginal maximum likelihood (MML) and standard variational inference (VI) is limited. We then apply the same approach to estimate mixture IRT models. In this case, AVI is computationally faster than MML estimation and standard VI. To demonstrate the practical applicability of our AVI approach, we use it to fit a seven-dimensional mixture IRT model to a narcissism inventory. Whereas quadrature-based methods cannot feasibly estimate models of this dimensionality, the efficient AVI approach even allows for computation of bootstrapped standard errors. We provide our code, along with an easy-to-use tool for fitting these models to new datasets.<\/jats:p>","DOI":"10.1017\/psy.2026.10105","type":"journal-article","created":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T07:12:12Z","timestamp":1778137932000},"page":"1479-1497","update-policy":"https:\/\/doi.org\/10.1017\/policypage","source":"Crossref","is-referenced-by-count":0,"title":["Estimating Discrete Latent Variable Models Using Amortized Variational Inference"],"prefix":"10.1017","volume":"91","author":[{"given":"Karel","family":"Veldkamp","sequence":"first","affiliation":[{"id":[{"id":"https:\/\/ror.org\/04dkp9463","id-type":"ROR","asserted-by":"publisher"}],"name":"Universiteit van Amsterdam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Raoul","family":"Grasman","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/04dkp9463","id-type":"ROR","asserted-by":"publisher"}],"name":"Universiteit van Amsterdam"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7168-3238","authenticated-orcid":false,"given":"Dylan","family":"Molenaar","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/04dkp9463","id-type":"ROR","asserted-by":"publisher"}],"name":"University of Amsterdam"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"56","published-online":{"date-parts":[[2026,5,7]]},"reference":[{"key":"S0033312326101057_r14","doi-asserted-by":"crossref","unstructured":"Debelak, R. , & Urban, C. J. (2021). An evaluation of deep learning approaches for factor analysis of response and response time data. Preprint.","DOI":"10.31234\/osf.io\/9rphk"},{"key":"S0033312326101057_r15","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-010-9161-9"},{"key":"S0033312326101057_r53","unstructured":"Vermunt, J. K. , & Magidson, J. (2021). Latent GOLD 6.0 user\u2019s guide [computer software manual]. https:\/\/www.statisticalinnovations.com"},{"key":"S0033312326101057_r7","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-009-9136-x"},{"key":"S0033312326101057_r44","unstructured":"Rainforth, T. , Kosiorek, A. , Le, T. A. , Maddison, C. , Igl, M. , Wood, F. , & Teh, Y. W. (2018). Tighter variational bounds are not necessarily better. In J. Dy & A. Krause (Eds.), International conference on machine learning (pp. 4277\u20134285)."},{"key":"S0033312326101057_r37","doi-asserted-by":"publisher","DOI":"10.1080\/00273171.2025.2456598"},{"key":"S0033312326101057_r32","first-page":"4189","article-title":"Estimating three-and four-parameter MIRT models with importance-weighted sampling enhanced variational auto-encoder","volume":"13","author":"Liu","year":"2022","journal-title":"Frontiers in Psychology"},{"key":"S0033312326101057_r23","doi-asserted-by":"publisher","DOI":"10.1051\/itmconf\/20246401003"},{"key":"S0033312326101057_r35","unstructured":"Maddison, C. J. , Mnih, A. , & Teh, Y. W. (2016). The concrete distribution: A continuous relaxation of discrete random variables. Preprint, arXiv:1611.00712."},{"key":"S0033312326101057_r49","unstructured":"Tucker, G. , Lawson, D. , Gu, S. , & Maddison, C. J. (2018). Doubly reparameterized gradient estimators for Monte Carlo objectives. Preprint, arXiv:1810.04152."},{"key":"S0033312326101057_r8","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-018-9646-5"},{"key":"S0033312326101057_r13","doi-asserted-by":"publisher","DOI":"10.1016\/j.pse.2014.11.001"},{"key":"S0033312326101057_r1","doi-asserted-by":"publisher","DOI":"10.1002\/9781119970583"},{"key":"S0033312326101057_r45","doi-asserted-by":"publisher","DOI":"10.1037\/0022-3514.54.5.890"},{"key":"S0033312326101057_r33","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-023-09939-0"},{"key":"S0033312326101057_r56","unstructured":"Wu, M. , Davis, R. L. , Domingue, B. W. , Piech, C. , & Goodman, N. (2021). Modeling item response theory with stochastic variational inference. Preprint, arXiv:2108.11579."},{"key":"S0033312326101057_r48","doi-asserted-by":"publisher","DOI":"10.1176\/ajp.155.10.1398"},{"key":"S0033312326101057_r38","doi-asserted-by":"publisher","DOI":"10.1007\/s11831-020-09422-4"},{"key":"S0033312326101057_r59","doi-asserted-by":"publisher","DOI":"10.3758\/s13428-024-02406-3"},{"key":"S0033312326101057_r18","doi-asserted-by":"publisher","DOI":"10.1086\/225676"},{"key":"S0033312326101057_r20","doi-asserted-by":"publisher","DOI":"10.1093\/pan\/mpq027"},{"key":"S0033312326101057_r27","article-title":"A systematic review of and reflection on the applications of factor mixture modeling","volume":"30","author":"Kim","year":"2023","journal-title":"Psychological Methods"},{"key":"S0033312326101057_r39","unstructured":"Muth\u00e9n, L. K. , & Muth\u00e9n, B. O. (2017). Mplus user\u2019s guide (8th ed.) [computer software manual]. Version 8."},{"key":"S0033312326101057_r3","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2017.1285773"},{"key":"S0033312326101057_r19","doi-asserted-by":"publisher","DOI":"10.1037\/a0038231"},{"key":"S0033312326101057_r40","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-024-00817-0"},{"key":"S0033312326101057_r58","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-020-09739-w"},{"key":"S0033312326101057_r5","unstructured":"Bouchattaoui, M. E. , Tami, M. , Lepetit, B. , & Courn\u00e8de, P.-H. (2023). Causal dynamic variational autoencoder for counterfactual regression in longitudinal data. Preprint, arXiv:2310.10559."},{"key":"S0033312326101057_r50","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-021-09748-3"},{"key":"S0033312326101057_r34","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v093.i14"},{"key":"S0033312326101057_r51","unstructured":"van den Oord, A. , Vinyals, O. , & Kavukcuoglu, K. (2017). Neural discrete representation learning. In I. Guyon, U. von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, & R. Garnett (Eds.), Advances in Neural Information Processing Systems (Vol. 30, pp. 6306\u20136315). Curran Associates, Inc."},{"key":"S0033312326101057_r57","doi-asserted-by":"publisher","DOI":"10.1086\/210470"},{"key":"S0033312326101057_r25","first-page":"91","article-title":"A specialized confirmatory mixture IRT modeling approach for multidimensional tests","volume":"61","author":"Jeon","year":"2019","journal-title":"Psychological Test and Assessment Modeling"},{"key":"S0033312326101057_r26","doi-asserted-by":"publisher","DOI":"10.1287\/opre.1.5.263"},{"key":"S0033312326101057_r42","first-page":"8024","volume-title":"Advances in neural information processing systems","volume":"32","author":"Paszke","year":"2019"},{"key":"S0033312326101057_r4","doi-asserted-by":"publisher","DOI":"10.1007\/BF02293801"},{"key":"S0033312326101057_r12","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-011-9207-7"},{"key":"S0033312326101057_r22","unstructured":"Hasan, M. , Deng, L. Y. , Sabatini, J. , Bowman, D. , Yang, C.-C. , & Hollander, J. (2022). Effect of Q-matrix misspecification on variational autoencoders (VAE) for multidimensional item response theory (MIRT) models estimation. In Antonija Mitrovic; Nigel Bosch (Eds). Proceedings of the 15th international conference on educational data mining (p. 811). International Educational Data Mining Society."},{"key":"S0033312326101057_r10","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-021-06005-7"},{"key":"S0033312326101057_r54","doi-asserted-by":"publisher","DOI":"10.3102\/1076998609346970"},{"key":"S0033312326101057_r36","volume-title":"Latent class analysis","author":"Magidson","year":"2020"},{"key":"S0033312326101057_r47","first-page":"177","article-title":"Applications of mixture IRT models: A literature review","volume":"17","author":"Sen","year":"2019","journal-title":"Measurement: Interdisciplinary Research and Perspectives"},{"key":"S0033312326101057_r9","doi-asserted-by":"publisher","DOI":"10.1111\/bmsp.12219"},{"key":"S0033312326101057_r52","doi-asserted-by":"publisher","DOI":"10.1111\/bmsp.12363"},{"key":"S0033312326101057_r31","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v042.i10"},{"key":"S0033312326101057_r2","unstructured":"Bengio, Y. , L\u00e9onard, N. , & Courville, A. (2013). Estimating or propagating gradients through stochastic neurons for conditional computation. Preprint, arXiv:1308.3432."},{"key":"S0033312326101057_r16","doi-asserted-by":"publisher","DOI":"10.1177\/0013164413494776"},{"key":"S0033312326101057_r24","unstructured":"Jang, E. , Gu, S. , & Poole, B. (2016). Categorical reparameterization with Gumbel-Softmax. Preprint, arXiv:1611.01144."},{"key":"S0033312326101057_r55","doi-asserted-by":"publisher","DOI":"10.1037\/pspp0000463"},{"key":"S0033312326101057_r41","doi-asserted-by":"publisher","DOI":"10.1007\/s11336-022-09884-4"},{"key":"S0033312326101057_r29","unstructured":"Lazarsfeld, P. F. (1950). The logical and mathematical foundation of latent structure analysis, Studies in Social Psychology in World War II Vol. IV: Measurement and Prediction (pp. 362\u2013412). Princeton University Press."},{"key":"S0033312326101057_r28","unstructured":"Kingma, D. P. , & Welling, M. (2013). Auto-encoding variational Bayes. Preprint, arXiv:1312.6114."},{"key":"S0033312326101057_r21","unstructured":"Gumbel, E. J. (1954). Statistical theory of extreme values and some practical applications, National Bureau of Standards Applied Mathematics Series, 33. U.S. Government Printing Office."},{"key":"S0033312326101057_r11","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN.2019.8852333"},{"key":"S0033312326101057_r17","doi-asserted-by":"publisher","DOI":"10.1177\/096228029600500205"},{"key":"S0033312326101057_r46","unstructured":"Reddi, S. J. , Kale, S. , & Kumar, S. (2019). On the convergence of Adam and beyond. Preprint, arXiv:1904.09237."},{"key":"S0033312326101057_r6","unstructured":"Burda, Y. , Grosse, R. , & Salakhutdinov, R. (2015). Importance weighted autoencoders. Preprint, arXiv:1509.00519."},{"key":"S0033312326101057_r30","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2510.17821"},{"key":"S0033312326101057_r43","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3042395"}],"container-title":["Psychometrika"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.cambridge.org\/core\/services\/aop-cambridge-core\/content\/view\/S0033312326101057","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T08:58:45Z","timestamp":1791277125000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.cambridge.org\/core\/product\/identifier\/S0033312326101057\/type\/journal_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,7]]},"references-count":59,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,9]]}},"alternative-id":["S0033312326101057"],"URL":"https:\/\/doi.org\/10.1017\/psy.2026.10105","relation":{},"ISSN":["0033-3123","1860-0980"],"issn-type":[{"value":"0033-3123","type":"print"},{"value":"1860-0980","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,7]]},"assertion":[{"value":"\u00a9 The Author(s), 2026. Published by Cambridge University Press on behalf of Psychometric Society","name":"copyright","label":"Copyright","group":{"name":"copyright_and_licensing","label":"Copyright and Licensing"}},{"value":"This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https:\/\/creativecommons.org\/licenses\/by\/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.","name":"license","label":"License","group":{"name":"copyright_and_licensing","label":"Copyright and Licensing"}},{"value":"This content has been made available to all.","name":"free","label":"Free to read"}]}}