{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T20:27:23Z","timestamp":1783542443559,"version":"3.55.0"},"reference-count":103,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62576266"],"award-info":[{"award-number":["62576266"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U21B2006"],"award-info":[{"award-number":["U21B2006"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["QTZX24003"],"award-info":[{"award-number":["QTZX24003"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["QTZX23018"],"award-info":[{"award-number":["QTZX23018"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100013314","name":"Higher Education Discipline Innovation Project","doi-asserted-by":"publisher","award":["B18039"],"award-info":[{"award-number":["B18039"]}],"id":[{"id":"10.13039\/501100013314","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62301384"],"award-info":[{"award-number":["62301384"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100010909","name":"Excellent Young Scientists Fund","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100010909","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,8]]},"DOI":"10.1109\/tpami.2026.3680869","type":"journal-article","created":{"date-parts":[[2026,4,6]],"date-time":"2026-04-06T19:56:01Z","timestamp":1775505361000},"page":"9675-9693","source":"Crossref","is-referenced-by-count":0,"title":["A Non-Negative Deep VAE: The Generalized Gamma Belief Network"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-3353-9975","authenticated-orcid":false,"given":"Zhibin","family":"Duan","sequence":"first","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-7911-7694","authenticated-orcid":false,"given":"Tiansheng","family":"Wen","sequence":"additional","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muyao","family":"Wang","sequence":"additional","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8875-1161","authenticated-orcid":false,"given":"Wenchao","family":"Chen","sequence":"additional","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hao","family":"Zhang","sequence":"additional","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5151-9388","authenticated-orcid":false,"given":"Bo","family":"Chen","sequence":"additional","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fei","family":"Meng","sequence":"additional","affiliation":[{"name":"Beijing Institute of Radio Measurement, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4046-163X","authenticated-orcid":false,"given":"Hongwei","family":"Liu","sequence":"additional","affiliation":[{"name":"National Key Lab of Radar Signal Processing, Xidian University, Xi’an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4253-2780","authenticated-orcid":false,"given":"Mingyuan","family":"Zhou","sequence":"additional","affiliation":[{"name":"McCombs School of Business, University of Texas, Austin, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Auto-encoding variational Bayes","author":"Kingma","year":"2013"},{"key":"ref2","first-page":"1278","article-title":"Stochastic backpropagation and approximate inference in deep generative models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Rezende"},{"key":"ref3","first-page":"19667","article-title":"NVAE: A deep hierarchical variational autoencoder","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Vahdat"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/k16-1002"},{"key":"ref5","article-title":"Variational graph auto-encoders","author":"Kipf","year":"2016"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2013.50"},{"key":"ref7","article-title":"Beta-VAE: Learning basic visual concepts with a constrained variational framework","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Higgins"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1017\/S0140525X16001837"},{"key":"ref9","article-title":"Autoencoding variational inference for topic models","author":"Srivastava","year":"2017"},{"key":"ref10","first-page":"4797","article-title":"Conditional image generation with PixelCNN decoders","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Van den Oord"},{"key":"ref11","article-title":"Very deep VAEs generalize autoregressive models and can outperform them on images","author":"Child","year":"2020"},{"key":"ref12","first-page":"2649","article-title":"Disentangling by factorising","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Kim"},{"key":"ref13","first-page":"2615","article-title":"Isolating sources of disentanglement in variational autoencoders","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref14","article-title":"Variational inference of disentangled latent concepts from unlabeled observations","author":"Kumar","year":"2017"},{"key":"ref15","first-page":"993","article-title":"Latent Dirichlet allocation","volume":"3","author":"Blei","year":"2003","journal-title":"J. Mach. Learn. Res."},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2013.211"},{"key":"ref17","first-page":"3043","article-title":"The Poisson gamma belief network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhou"},{"issue":"1","key":"ref18","first-page":"5656","article-title":"Augmentable gamma belief networks","volume":"17","author":"Zhou","year":"2016","journal-title":"J. Mach. Learn. Res."},{"key":"ref19","first-page":"864","article-title":"Deep latent Dirichlet allocation with topic-layer-adaptive stochastic gradient Riemannian MCMC","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Cong"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1038\/44565"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.3003660"},{"key":"ref22","article-title":"Whai: Weibull hybrid autoencoding inference for deep topic modeling","author":"Zhang","year":"2018"},{"key":"ref23","first-page":"2903","article-title":"Sawtooth factorial topic embeddings guided gamma belief network","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Duan"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1639"},{"key":"ref25","first-page":"8731","article-title":"Bayesian progressive deep topic model with knowledge informed textual data coarsening process","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Duan"},{"key":"ref26","first-page":"488","article-title":"Deep relational topic modeling via graph Poisson gamma belief network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wang"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3192319"},{"key":"ref28","first-page":"3745","article-title":"Ladder variational autoencoders","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"S\u00f8nderby"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1111\/insr.12502"},{"key":"ref30","volume-title":"Pattern Recognition and Machine Learning","volume":"4","author":"Bishop","year":"2006"},{"key":"ref31","first-page":"690","article-title":"Variational sparse coding","volume-title":"Proc. Uncertainty Artif. Intell.","author":"Tonolini"},{"key":"ref32","first-page":"4402","article-title":"Disentangling disentanglement in variational autoencoders","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Mathieu"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1194"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.5555\/3104322.3104425"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.10779"},{"key":"ref36","first-page":"4117","article-title":"Hierarchical VAEs know what they don\u2019t know","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Havtorn"},{"key":"ref37","first-page":"6551","article-title":"BIVA: A very deep hierarchy of latent variables for generative modeling","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Maal\u00f8e"},{"key":"ref38","first-page":"2397","article-title":"Avoiding latent variable collapse with generative skip models","volume-title":"Proc. 22nd Int. Conf. Artif. Intell. Statist.","author":"Dieng"},{"key":"ref39","article-title":"Deep attentive variational inference","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Apostolopoulou"},{"key":"ref40","article-title":"Understanding disentangling in BETA-VAE","author":"Burgess","year":"2018"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107514"},{"key":"ref42","article-title":"Stick-breaking variational autoencoders","author":"Nalisnick","year":"2016"},{"key":"ref43","first-page":"8451","article-title":"Deep Poisson gamma dynamical systems","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Guo"},{"key":"ref44","first-page":"6515","article-title":"Convolutional Poisson gamma belief network","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Wang"},{"key":"ref45","first-page":"801","article-title":"Efficient sparse coding algorithms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Lee"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.5555\/3104322.3104374"},{"key":"ref47","first-page":"10534","article-title":"The sparse manifold transform","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1167\/jov.20.12.10"},{"key":"ref49","first-page":"2110","article-title":"Adversarial robustness of supervised sparse coding","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Sulam"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_01715"},{"key":"ref51","article-title":"K-sparse autoencoders","author":"Makhzani","year":"2013"},{"key":"ref52","article-title":"Disentangling dense embeddings with sparse autoencoders","author":"O\u2019Neill","year":"2024"},{"key":"ref53","article-title":"Sparse autoencoders enable scalable and reliable circuit identification in language models","author":"O\u2019Neill","year":"2024"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2024.111187"},{"key":"ref55","article-title":"Improved training of sparse coding variational autoencoder via weight normalization","author":"Jiang","year":"2021"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1198\/jcgs.2010.09014"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref59","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"ref60","article-title":"Preventing posterior collapse with delta-VAEs","author":"Razavi","year":"2019"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1016\/S0893-6080(98)00140-3"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref63","first-page":"2526","article-title":"One-shot learning by inverting a compositional causal process","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Lake"},{"key":"ref64","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.425"},{"key":"ref66","first-page":"1558","article-title":"Autoencoding beyond pixels using a learned similarity metric","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Larsen"},{"key":"ref67","article-title":"PixelCNN++: Improving the pixelCNN with discretized logistic mixture likelihood and other modifications","author":"Salimans","year":"2017"},{"key":"ref68","article-title":"Density estimation using real NVP","author":"Dinh","year":"2016"},{"key":"ref69","first-page":"10236","article-title":"Glow: Generative flow with invertible 1 \u00d7 1 convolutions","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Kingma"},{"key":"ref70","first-page":"2722","article-title":"Flow++: Improving flow-based generative models with variational dequantization and architecture design","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Ho"},{"key":"ref71","first-page":"1747","article-title":"Pixel recurrent neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"van Den"},{"key":"ref72","first-page":"4055","article-title":"Image transformer","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Parmar"},{"key":"ref73","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ho"},{"key":"ref74","article-title":"Score-based generative modeling through stochastic differential equations","author":"Song","year":"2020"},{"key":"ref75","first-page":"5035","article-title":"DVAE++: Discrete variational autoencoders with overlapping transformations","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Vahdat"},{"key":"ref76","first-page":"3549","article-title":"Towards conceptual compression","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Gregor"},{"key":"ref77","first-page":"1660","article-title":"VFlow: More expressive generative flows with variational data augmentation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chen"},{"key":"ref78","article-title":"Augmented normalizing flows: Bridging the gap between generative flows and latent variable models","author":"Huang","year":"2020"},{"key":"ref79","first-page":"9913","article-title":"Residual flows for invertible generative modeling","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref80","article-title":"PixelVAE++: Improved pixelvae with discrete prior","author":"Sadeghi","year":"2019"},{"key":"ref81","first-page":"1214","article-title":"VAE with a vampprior","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Tomczak"},{"key":"ref82","article-title":"MAE: Mutual posterior-divergence regularization for variational autoencoders","author":"Ma","year":"2019"},{"key":"ref83","article-title":"Variational lossy autoencoder","author":"Chen","year":"2016"},{"key":"ref84","first-page":"5891","article-title":"MaCow: Masked convolutional generative flow","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ma"},{"key":"ref85","first-page":"864","article-title":"PixelSnail: An improved autoregressive generative model","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chen"},{"key":"ref86","first-page":"8162","article-title":"Improved denoising diffusion probabilistic models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Nichol"},{"key":"ref87","article-title":"A study on ReLU and softmax in transformer","author":"Shen","year":"2023"},{"key":"ref88","article-title":"dSprites: Disentanglement testing sprites dataset","author":"Matthey","year":"2020"},{"key":"ref89","article-title":"3D shapes dataset","author":"Burgess","year":"2018"},{"key":"ref90","first-page":"1252","article-title":"Deep visual analogy-making","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Reed"},{"key":"ref91","first-page":"185","article-title":"Learning deep disentangled embeddings with the F-statistic loss","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ridgeway"},{"key":"ref92","article-title":"A framework for the quantitative evaluation of disentangled representations","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Eastwood"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1147\/rd.41.0066"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2010.2068870"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1007\/s10463-011-0343-8"},{"key":"ref96","first-page":"4114","article-title":"Challenging common assumptions in the unsupervised learning of disentangled representations","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Locatello"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00325"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1042"},{"key":"ref99","doi-asserted-by":"publisher","DOI":"10.1016\/B978-1-55860-377-6.50048-7"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-28997-2_32"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1145\/219717.219748"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1162"},{"key":"ref103","article-title":"Unsupervised feature learning and deep learning: A review and new perspectives","author":"Bengio","year":"2012"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11595778\/11474987.pdf?arnumber=11474987","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T19:43:03Z","timestamp":1783539783000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11474987\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8]]},"references-count":103,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2026.3680869","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8]]}}}