{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,2]],"date-time":"2026-08-02T18:06:46Z","timestamp":1785694006091,"version":"3.56.0"},"reference-count":265,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"11","license":[{"start":{"date-parts":[[2022,11,1]],"date-time":"2022-11-01T00:00:00Z","timestamp":1667260800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"MRC Innovation Fellowship","award":["MR\/S003916\/1"],"award-info":[{"award-number":["MR\/S003916\/1"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2022,11,1]]},"DOI":"10.1109\/tpami.2021.3116668","type":"journal-article","created":{"date-parts":[[2021,9,30]],"date-time":"2021-09-30T21:12:05Z","timestamp":1633036325000},"page":"7327-7347","source":"Crossref","is-referenced-by-count":540,"title":["Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models"],"prefix":"10.1109","volume":"44","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1538-7909","authenticated-orcid":false,"given":"Sam","family":"Bond-Taylor","sequence":"first","affiliation":[{"name":"Department of Computer Science, Durham University, Durham, U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8805-3921","authenticated-orcid":false,"given":"Adam","family":"Leach","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Durham University, Durham, U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2445-6112","authenticated-orcid":false,"given":"Yang","family":"Long","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Durham University, Durham, U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6821-3924","authenticated-orcid":false,"given":"Chris G.","family":"Willcocks","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Durham University, Durham, U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1093\/imaiai\/iaw003"},{"key":"ref2","first-page":"159","article-title":"Fixing a broken ELBO","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Alemi"},{"key":"ref3","article-title":"Generalized energy based models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Arbel"},{"key":"ref4","article-title":"Towards principled methods for training generative adversarial networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Arjovsky"},{"key":"ref5","first-page":"214","article-title":"Wasserstein generative adversarial networks","author":"Arjovsky","year":"2017","journal-title":"Proc. Int. Conf. Mach. Learn."},{"key":"ref6","article-title":"Stochastic variational video prediction","volume-title":"Proc. 6th Int. Conf. Learn. Representations","author":"Babaeizadeh"},{"key":"ref7","first-page":"66","article-title":"Resampled priors for variational autoencoders","volume-title":"Proc. 22nd Int. Conf. Artif. Intell. Statist.","author":"Bauer"},{"key":"ref8","first-page":"573","article-title":"Invertible residual networks","volume-title":"Proc. 36th Int. Conf. Mach. Learn.","author":"Behrmann"},{"key":"ref9","first-page":"1137","article-title":"A neural probabilistic language model","volume":"3","author":"Bengio","year":"2003","journal-title":"J. Mach. Learn. Res."},{"key":"ref10","article-title":"Estimating or propagating gradients through stochastic neurons for conditional computation","author":"Bengio","year":"2013"},{"key":"ref11","first-page":"899","article-title":"Generalized denoising auto-encoders as generative models","volume-title":"Proc. 26th Int. Conf. Neural Inf. Process. Syst.","author":"Bengio"},{"key":"ref12","article-title":"IDF++: Analyzing and improving integer discrete flows for lossless compression","volume-title":"Proc. Int. Conf. Learn. Representations","author":"van den Berg"},{"key":"ref13","first-page":"393","article-title":"Sylvester normalizing flows for variational inference","volume-title":"Proc. 34th Conf. Uncertainty Artif. Intell.","author":"van den Berg"},{"key":"ref14","article-title":"Demystifying MMD GANs","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Bi\u0144kowski"},{"key":"ref15","first-page":"600","article-title":"Optimizing the latent space of generative networks","author":"Bojanowski","year":"2018","journal-title":"Proc. Int. Conf. Mach. Learn."},{"key":"ref16","article-title":"Gradient origin networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Bond-Taylor"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/K16-1002"},{"key":"ref18","article-title":"Large scale GAN training for high fidelity natural image synthesis","author":"Brock","year":"2018","journal-title":"Proc. Int. Conf. Learn. Representations"},{"key":"ref19","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Proc. Adv. Neural Inf. Process. Syst."},{"key":"ref20","article-title":"Importance weighted autoencoders","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Burda"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1090\/conm\/226"},{"key":"ref22","first-page":"33","article-title":"On contrastive divergence learning","volume-title":"Proc. 10th Int. Workshop Artif. Intell. Statist.","author":"Carreira-Perpinan"},{"key":"ref23","article-title":"Your GAN is secretly an energy-based model and you should use discriminator driven latent sampling","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Che"},{"key":"ref24","first-page":"1660","article-title":"VFlow: More expressive generative flows with variational data augmentation","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","author":"Chen"},{"key":"ref25","first-page":"9916","article-title":"Residual flows for invertible generative modeling","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-04167-0"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01243"},{"key":"ref28","article-title":"Variational lossy autoencoder","volume-title":"Proc. 5th Int. Conf. Learn. Representations","author":"Chen"},{"key":"ref29","first-page":"864","article-title":"PixelSNAIL: An improved autoregressive generative model","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Chen"},{"key":"ref30","article-title":"Very deep VAEs generalize autoregressive models and can outperform them on images","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Child"},{"key":"ref31","article-title":"Generating long sequences with sparse transformers","author":"Child","year":"2019"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.2307\/j.ctv36zrf8.5"},{"key":"ref34","article-title":"Empirical evaluation of gated recurrent neural networks on sequence modeling","author":"Chung","year":"2014","journal-title":"Proc. NIPS 2014 Workshop Deep Learn."},{"key":"ref35","first-page":"2133","article-title":"Relaxing bijectivity constraints with continuously indexed normalising flows","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Cornish"},{"key":"ref36","first-page":"1078","article-title":"Inference suboptimality in variational autoencoders","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Cremer"},{"key":"ref37","article-title":"Diagnosing and enhancing VAE models","author":"Dai","year":"2018","journal-title":"Proc. Int. Conf. Learn. Representations"},{"key":"ref38","first-page":"1263","article-title":"Block neural autoregressive flow","volume-title":"Proc. 35th Conf. Uncertainty Artif. Intell.","author":"Cao"},{"key":"ref39","first-page":"1486","article-title":"Deep generative image models using a Laplacian pyramid of adversarial networks","volume-title":"Proc. 28th Int. Conf. Neural Inf. Process. Syst.","author":"Denton"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/vl\/N19-142"},{"key":"ref41","article-title":"NICE: Non-linear independent components estimation","volume-title":"Proc. 3rd Int. Conf. Learn. Representations Workshop","author":"Dinh"},{"key":"ref42","article-title":"Density estimation using Real NVP","volume-title":"Proc. 5th Int. Conf. Learn. Representations","author":"Dinh"},{"key":"ref43","article-title":"A RAD approach to deep mixture models","volume-title":"Proc. Int. Conf. Learn. Representations Workshop","author":"Dinh"},{"key":"ref44","first-page":"658","article-title":"Generating images with perceptual similarity metrics based on deep networks","volume-title":"Proc. 30th Int. Conf. Neural Inf. Process. Syst.","author":"Dosovitskiy"},{"key":"ref45","first-page":"3608","article-title":"Implicit generation and generalization in energy-based models","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Du"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2021.2003201"},{"key":"ref47","first-page":"3140","article-title":"Augmented neural ODEs","volume-title":"Proc. 32nd Int. Conf. Neural Inf. Process. Syst.","author":"Dupont"},{"key":"ref48","article-title":"Generative models as distributions of functions","author":"Dupont","year":"2021"},{"key":"ref49","article-title":"Cubic-spline flows","volume-title":"Proc. Int. Conf. Mach. Learn. Workshop","author":"Durkan"},{"key":"ref50","first-page":"7511","article-title":"Neural spline flows","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Durkan"},{"key":"ref51","first-page":"2771","article-title":"On contrastive learning for likelihood-free inference","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","author":"Durkan"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01268"},{"key":"ref53","first-page":"3154","article-title":"How to train your neural ODE: The world of jacobian and kinetic regularization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Finlay"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00754"},{"key":"ref55","article-title":"Learning energy-based models by diffusion recovery likelihood","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Gao"},{"key":"ref56","first-page":"881","article-title":"MADE: Masked autoencoder for distribution estimation","volume-title":"Proc. 32nd Int. Conf. Mach. Learn.","author":"Germain"},{"key":"ref57","article-title":"From variational to deterministic autoencoders","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Ghosh"},{"key":"ref58","first-page":"2672","article-title":"Generative adversarial nets","volume-title":"Proc. 27th Int. Conf. Neural Inf. Process. Syst.","author":"Goodfellow"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-020-05929-w"},{"key":"ref60","article-title":"Exposing the implicit energy networks behind masked language models via metropolis\u2013hastings","author":"Goyal","year":"2021"},{"key":"ref61","article-title":"FFJORD: Free-form continuous dynamics for scalable reversible generative models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Grathwohl"},{"key":"ref62","first-page":"3831","article-title":"Oops I took a gradient: Scalable sampling for discrete distributions","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Grathwohl"},{"key":"ref63","first-page":"3732","article-title":"Learning the stein discrepancy for training and evaluating energy-based models without sampling","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Grathwohl"},{"key":"ref64","article-title":"Your classifier is secretly an energy based model and you should treat it like one","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Grathwohl"},{"key":"ref65","article-title":"No MCMC for me: Amortized sampling for fast and stable training of energy-based models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Grathwohl"},{"key":"ref66","first-page":"1462","article-title":"DRAW: A recurrent neural network for image generation","volume-title":"Proc. 32nd Int. Conf. Mach. Learn.","author":"Gregor"},{"key":"ref67","article-title":"Generative minimization networks: Training GANs without competition","author":"Grnarova","year":"2021"},{"key":"ref68","first-page":"12092","article-title":"A domain agnostic measure for monitoring and evaluating GANs","author":"Grnarova","year":"2019","journal-title":"Prc. 33rd Int. Conf. Neural Inf. Process. Syst."},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11829"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2021.3130191"},{"key":"ref71","first-page":"5769","article-title":"Improved training of wasserstein GANs","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Gulrajani"},{"key":"ref72","article-title":"PixelVAE: A latent variable model for natural images","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Gulrajani"},{"key":"ref73","article-title":"Towards GAN benchmarks which require generalization","volume-title":"Proc. 7th Int. Conf. Learn. Representations","author":"Gulrajani"},{"key":"ref74","first-page":"297","article-title":"Noise-contrastive estimation: A new estimation principle for unnormalized statistical models","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Gutmann"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-31635-3_22"},{"key":"ref76","article-title":"Dream to control: Learning behaviors by latent imagination","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Hafner"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.10902"},{"key":"ref78","article-title":"Variational inference with orthogonal normalizing flows","volume-title":"Proc. Workshop Bayesian Deep Learn. Neural Inf. Process. Syst.","author":"Hasenclever"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref80","first-page":"6629","article-title":"GANs trained by a two time-scale update rule converge to a local Nash equilibrium","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Heusel"},{"key":"ref81","article-title":"beta-VAE: Learning basic visual concepts with a constrained variational framework","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Higgins"},{"key":"ref82","first-page":"448","article-title":"Optimal perceptual inference","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Hinton"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1162\/089976602760128018"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1162\/neco.2006.18.7.1527"},{"key":"ref85","first-page":"2722","article-title":"Flow++: Improving flow-based generative models with variational dequantization and architecture design","volume-title":"Proc. 36th Int. Conf. Mach. Learn.","author":"Ho"},{"key":"ref86","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho","year":"2020","journal-title":"Proc. Adv. Neural Inf. Process. Syst."},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref88","article-title":"NeuTra-lizing bad geometry in hamiltonian Monte Carlo using neural transport","author":"Hoffman","year":"2019"},{"key":"ref89","article-title":"ELBO surgery: Yet another way to carve up the variational evidence lower bound","volume-title":"Proc. Conf. Neural Inf. Process. Syst. Workshop","author":"Hoffman"},{"key":"ref90","first-page":"2771","article-title":"Emerging convolutions for generative normalizing flows","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Hoogeboom"},{"key":"ref91","first-page":"18249","article-title":"The convolution exponential and generalized sylvester flows","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Hoogeboom"},{"key":"ref92","article-title":"Argmax flows and multinomial diffusion: Learning categorical distributions","author":"Hoogeboom","year":"2021"},{"key":"ref93","first-page":"12134","article-title":"Integer discrete flows and lossless compression","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Hoogeboom"},{"key":"ref94","article-title":"Augmented normalizing flows: Bridging the gap between generative flows and latent variable models","author":"Huang","year":"2020","journal-title":"Workshop Integr. Deep Neural Models Differ. Equ."},{"key":"ref95","first-page":"2078","article-title":"Neural autoregressive flows","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Huang"},{"key":"ref96","article-title":"Learnable explicit density for continuous latent space and variational inference","author":"Huang","year":"2017"},{"key":"ref97","first-page":"52","article-title":"IntroVAE: Introspective variational autoencoders for photographic image synthesis","volume-title":"Proc. 32nd Int. Conf. Neural Inf. Process. Syst.","author":"Huang"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.202"},{"key":"ref99","article-title":"How (not) to train your generative model: Scheduled sampling, likelihood, adversary?","author":"Husz\u00e1r","year":"2015"},{"key":"ref100","first-page":"695","article-title":"Estimation of non-normalized statistical models by score matching","volume":"6","author":"Hyv\u00e4rinen","year":"2005","journal-title":"J. Mach. Learn. Res."},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1007\/s13398-014-0173-7.2"},{"key":"ref102","first-page":"3009","article-title":"Sum-of-squares polynomial flow","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Jaini"},{"key":"ref103","article-title":"Categorical reparameterization with Gumbel-Softmax","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Jang"},{"key":"ref104","article-title":"TransGAN: Two transformers can make one strong GAN","author":"Jiang","year":"2021"},{"key":"ref105","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.2992393"},{"key":"ref106","article-title":"The relativistic discriminator: A key element missing from standard GAN","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Jolicoeur-Martineau"},{"key":"ref107","article-title":"Gotta go fast when generating data with score-based models","author":"Jolicoeur-Martineau","year":"2021"},{"key":"ref108","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007665907178"},{"key":"ref109","first-page":"5006","article-title":"Distribution augmentation for generative modeling","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","author":"Jun"},{"key":"ref110","first-page":"2390","article-title":"Fast decoding in sequence models using discrete latent variables","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Kaiser"},{"key":"ref111","first-page":"5635","article-title":"Invertible convolutional flow","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Karami"},{"key":"ref112","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00782"},{"key":"ref113","article-title":"Progressive growing of GANs for improved quality, stability, and variation","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Karras"},{"key":"ref114","first-page":"12104","article-title":"Training generative adversarial networks with limited data","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Karras"},{"key":"ref115","article-title":"Alias-free generative adversarial networks","author":"Karras","year":"2021"},{"key":"ref116","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00453"},{"key":"ref117","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00813"},{"key":"ref118","first-page":"5156","article-title":"Transformers are RNNs: Fast autoregressive transformers with linear attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Katharopoulos"},{"key":"ref119","first-page":"5562","article-title":"The Lipschitz constant of self-attention","author":"Kim","year":"2021","journal-title":"Proc. Int. Conf. Mach. Learn."},{"key":"ref120","article-title":"Deep directed generative models with energy-based probability estimation","author":"Kim","year":"2016"},{"key":"ref121","first-page":"2678","article-title":"Semi-amortized variational autoencoders","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Kim"},{"key":"ref122","article-title":"Auto-encoding variational Bayes","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Kingma"},{"key":"ref123","first-page":"10215","article-title":"Glow: Generative flow with invertible 1x1 convolutions","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Kingma"},{"key":"ref124","first-page":"4743","article-title":"Improved variational inference with inverse autoregressive flow","volume-title":"Proc. 30th Int. Conf. Neural Inf. Process. Syst.","author":"Kingma"},{"key":"ref125","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.2992934"},{"key":"ref126","article-title":"Learning multiple layers of features from tiny images","volume-title":"Tech. Rep. Krizhevsky_2009_17719","author":"Krizhevsky","year":"2009"},{"key":"ref127","first-page":"14910","article-title":"MelGAN: Generative adversarial networks for conditional waveform synthesis","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Kumar"},{"key":"ref128","article-title":"VideoFlow: A flow-based generative model for video","volume-title":"Proc. Int. Conf. Mach. Learn. Workshop","author":"Kumar"},{"key":"ref129","article-title":"Maximum entropy generators for energy-based models","author":"Kumar","year":"2019"},{"key":"ref130","first-page":"29","article-title":"The neural autoregressive distribution estimator","volume-title":"Proc. 14th Int. Conf. Artif. Intell. Statist.","author":"Larochelle"},{"key":"ref131","first-page":"1558","article-title":"Autoencoding beyond pixels using a learned similarity metric","volume-title":"Proc. 33rd Int. Conf. Mach. Learn.","author":"Larsen"},{"key":"ref132","article-title":"A tutorial on energy-based learning","volume-title":"Predicting Structured Data 1","author":"LeCun","year":"2006"},{"key":"ref133","article-title":"Stochastic adversarial video prediction","author":"Lee","year":"2018"},{"key":"ref134","doi-asserted-by":"publisher","DOI":"10.3390\/e23030269"},{"key":"ref135","article-title":"Geometric GAN","author":"Lim","year":"2017"},{"key":"ref136","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01474"},{"key":"ref137","article-title":"Towards faster and stabilized GAN training for high-fidelity few-shot image synthesis","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Liu"},{"key":"ref138","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00648"},{"key":"ref139","first-page":"700","article-title":"Unsupervised image-to-image translation networks","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref140","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01149"},{"key":"ref141","first-page":"13266","article-title":"The continuous Bernoulli: Fixing a pervasive error in variational autoencoders","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Loaiza-Ganem"},{"key":"ref142","first-page":"13266","article-title":"Implicit normalizing flows","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Lu"},{"key":"ref143","first-page":"698","article-title":"Are GANs created equal? A large-scale study","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Lucic"},{"key":"ref144","first-page":"5893","article-title":"MaCow: Masked convolutional generative flow","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Ma"},{"key":"ref145","article-title":"MAE: Mutual posterior-divergence regularization for variational AutoEncoders","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Ma"},{"key":"ref146","first-page":"6551","article-title":"BIVA: A very deep hierarchy of latent variables for generative modeling","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Maal\u00f8e"},{"key":"ref147","article-title":"The concrete distribution: A continuous relaxation of discrete random variables","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Maddison"},{"key":"ref148","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.06083"},{"key":"ref149","article-title":"Adversarial autoencoders","author":"Makhzani","year":"2016"},{"key":"ref150","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.304"},{"key":"ref151","article-title":"SampleRNN: An unconditional end-to-end neural audio generation model","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Mehri"},{"key":"ref152","article-title":"Improved autoregressive modeling with distribution smoothing","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Meng"},{"key":"ref153","first-page":"6673","article-title":"Autoregressive score matching","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Meng"},{"key":"ref154","article-title":"Generating high fidelity images with subscale pixel networks and multidimensional upscaling","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Menick"},{"key":"ref155","article-title":"Conditional generative adversarial nets","author":"Mirza","year":"2014"},{"key":"ref156","article-title":"Spectral normalization for generative adversarial networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Miyato"},{"key":"ref157","doi-asserted-by":"publisher","DOI":"10.1145\/3341156"},{"key":"ref158","first-page":"1735","article-title":"Autoregressive energy machines","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Nash"},{"key":"ref159","first-page":"13955","article-title":"The implicit Metropolis-Hastings algorithm","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Neklyudov"},{"key":"ref160","doi-asserted-by":"publisher","DOI":"10.1109\/RoboMech.2019.8704766"},{"key":"ref161","article-title":"Improved denoising diffusion probabilistic models","author":"Nichol","year":"2021"},{"key":"ref162","article-title":"RelGAN: Relational generative adversarial networks for text generation","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Nie"},{"key":"ref163","first-page":"12685","article-title":"SurVAE flows: Surjections to bridge the gap between VAEs and flows","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Nielsen"},{"key":"ref164","article-title":"Learning energy-based model with flow-based backbone by neural transport MCMC","author":"Nijkamp","year":"2020"},{"key":"ref165","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5973"},{"key":"ref166","first-page":"5232","article-title":"Learning non-convergent non-persistent short-run MCMC toward energy-based model","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Nijkamp"},{"key":"ref167","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58539-6_22"},{"key":"ref168","first-page":"8753","article-title":"Exemplar VAE: Linking generative models, nearest neighbor retrieval, and data augmentation","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Norouzi"},{"key":"ref169","first-page":"271","article-title":"f-GAN: Training generative neural samplers using variational divergence minimization","volume-title":"Proc. 30th Int. Conf. Neural Inf. Process. Syst.","author":"Nowozin"},{"key":"ref170","first-page":"2642","article-title":"Conditional image synthesis with auxiliary classifier GANs","volume-title":"Proc. 34th Int. Conf. Mach. Learn.","author":"Odena"},{"key":"ref171","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i10.17113"},{"key":"ref172","first-page":"3936","article-title":"Autoregressive Quantile networks for generative modeling","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Ostrovski"},{"key":"ref173","doi-asserted-by":"publisher","DOI":"10.1109\/ISACV.2018.8354080"},{"key":"ref174","article-title":"Efficient learning of generative models via finite-difference score matching,","volume":"34","author":"Pang","year":"2020","journal-title":"Proc. Int. Conf. Neural Inf. Process. Syst."},{"issue":"57","key":"ref175","first-page":"1","article-title":"Normalizing flows for probabilistic modeling and inference","volume":"22","author":"Papamakarios","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"ref176","first-page":"2335","article-title":"Masked autoregressive flow for density estimation","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Papamakarios"},{"key":"ref177","first-page":"4055","article-title":"Image transformer","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Parmar"},{"key":"ref178","doi-asserted-by":"publisher","DOI":"10.1201\/9780203749319"},{"key":"ref179","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683143"},{"key":"ref180","article-title":"Unsupervised representation learning with deep convolutional generative adversarial networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Radford"},{"key":"ref181","first-page":"8821","article-title":"Zero-shot text-to-image generation","author":"Ramesh","year":"2021","journal-title":"Proc. 38th Int. Conf. Mach. Learn."},{"key":"ref182","first-page":"14837","article-title":"Generating diverse high-fidelity images with VQ-VAE-2","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Razavi"},{"key":"ref183","first-page":"2912","article-title":"Parallel multiscale autoregressive density estimation","volume-title":"Proc. 34th Int. Conf. Mach. Learn.","author":"Reed"},{"key":"ref184","first-page":"1530","article-title":"Variational inference with normalizing flows","volume-title":"Proc. 32nd Int. Conf. Mach. Learn.","author":"Rezende"},{"key":"ref185","first-page":"1278","article-title":"Stochastic backpropagation and approximate inference in deep generative models","volume-title":"Proc. 31st Int. Conf. Mach. Learn.","author":"Rezende"},{"key":"ref186","doi-asserted-by":"publisher","DOI":"10.2307\/3318418"},{"key":"ref187","first-page":"2234","article-title":"Improved techniques for training GANs","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Salimans"},{"key":"ref188","article-title":"PixelCNN++: Improving the PixelCNN with discretized logistic mixture likelihood and other modifications","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Salimans"},{"key":"ref189","first-page":"1218","article-title":"Markov chain Monte Carlo and variational inference: Bridging the gap","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Salimans"},{"key":"ref190","article-title":"Deep energy estimator networks","author":"Saremi","year":"2018"},{"key":"ref191","article-title":"UNIT-DDPM: UNpaired image translation with denoising diffusion probabilistic models","author":"Sasaki","year":"2021"},{"key":"ref192","article-title":"Making the world differentiable: On using self-supervised fully recurrent neural networks for dynamic reinforcement learning and planning in non-stationary environments","author":"Schmidhuber","year":"1990","journal-title":"Tech. Rep. FKI-126\u201390"},{"key":"ref193","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2020.04.008"},{"key":"ref194","first-page":"4644","article-title":"A spectral approach to gradient estimation for implicit distributions","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Shi"},{"key":"ref195","first-page":"9005","article-title":"Small-GAN: Speeding up GAN training using core-sets","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","author":"Sinha"},{"key":"ref196","first-page":"7462","article-title":"Implicit neural representations with periodic activation functions","volume":"33","author":"Sitzmann","year":"2020","journal-title":"Proc. Neural Inf. Process. Syst."},{"key":"ref197","first-page":"2256","article-title":"Deep Unsupervised Learning using Nonequilibrium Thermodynamics","volume-title":"Proc. 32nd Int. Conf. Mach. Learn.","author":"Sohl-Dickstein"},{"key":"ref198","article-title":"Denoising diffusion implicit models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Song"},{"key":"ref199","first-page":"5140","article-title":"A-NICE-MC: Adversarial training for MCMC","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Song"},{"key":"ref200","first-page":"11895","article-title":"Generative modeling by estimating gradients of the data distribution","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Song"},{"key":"ref201","first-page":"574","article-title":"Sliced score matching: A scalable approach to density and score estimation","volume-title":"Proc. Conf. Uncertainty Artif. Intell.","author":"Song"},{"key":"ref202","article-title":"How to train your energy-based models","author":"Song","year":"2021"},{"key":"ref203","first-page":"11004","article-title":"MintNet: Building invertible neural networks with masked convolutions","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Song"},{"key":"ref204","article-title":"Score-based generative modeling through stochastic differential equations","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Song"},{"key":"ref205","article-title":"Discriminator contrastive divergence: Semi-amortized generative modeling by exploring energy of the discriminator","author":"Song","year":"2020"},{"key":"ref206","first-page":"789","article-title":"On the convergence properties of contrastive divergence","volume-title":"Proc. 13th Int. Conf. Artif. Intell. Statist.","author":"Sutskever"},{"key":"ref207","first-page":"1201","article-title":"On autoencoders and score matching for energy based models","volume-title":"Proc. 28th Int. Conf. Mach. Learn.","author":"Swersky"},{"key":"ref208","article-title":"Amortised MAP inference for image super-resolution","volume-title":"Proc. Int. Conf. Learn. Representations","author":"S\u00f8nderby"},{"key":"ref209","first-page":"3745","article-title":"Ladder variational autoencoders","volume-title":"Proc. 30th Int. Conf. Neural Inf. Process. Syst.","author":"S\u00f8nderby"},{"key":"ref210","first-page":"7537","article-title":"Fourier features let networks learn high frequency functions in low dimensional domains","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Tancik"},{"key":"ref211","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN48605.2020.9207181"},{"key":"ref212","first-page":"1927","article-title":"Generative image modeling using spatial LSTMs","volume-title":"Proc. 28th Conf. Neural Inf. Process. Syst.","author":"Theis"},{"key":"ref213","article-title":"A note on the evaluation of generative models","author":"Theis","year":"2016","journal-title":"Proc. Int. Conf. Learn. Representations"},{"key":"ref214","doi-asserted-by":"publisher","DOI":"10.1145\/1390156.1390290"},{"key":"ref215","first-page":"15730","article-title":"Gradient-based adaptive Markov chain Monte Carlo","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Titsias"},{"key":"ref216","article-title":"Wasserstein auto-encoders","author":"Tolstikhin","year":"2018","journal-title":"Int. Conf. Learn. Representations"},{"key":"ref217","first-page":"1214","article-title":"VAE with a VampPrior","volume-title":"Proc. 21st Int. Conf. Artif. Intell. Statist.","author":"Tomczak"},{"key":"ref218","article-title":"Variational Gaussian process","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Tran"},{"key":"ref219","first-page":"14719","article-title":"Discrete flows: Invertible generative models of discrete data","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Tran"},{"key":"ref220","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3049346"},{"key":"ref221","first-page":"13253","article-title":"Self-supervised GAN: Analysis and improvement with multi-class minimax game","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Tran"},{"key":"ref222","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511984679.006"},{"key":"ref223","first-page":"6345","article-title":"Metropolis-Hastings generative adversarial networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Turner"},{"key":"ref224","first-page":"3084","article-title":"Theoretical guarantees for sampling and inference in generative models with latent diffusions","volume-title":"Proc. Annu. Conf. Learn. Theory","author":"Tzen"},{"key":"ref225","first-page":"2175","article-title":"RNADE: The real-valued neural autoregressive density-estimator","volume-title":"Proc. 26th Int. Conf. Neural Inf. Process. Syst.","author":"Uria"},{"key":"ref226","first-page":"19667","article-title":"NVAE: A deep hierarchical variational autoencoder","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Vahdat"},{"key":"ref227","article-title":"Score-based generative modeling in latent space","author":"Vahdat","year":"2021"},{"key":"ref228","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2021.3069634"},{"key":"ref229","article-title":"WaveNet: A generative model for raw audio","author":"van den Oord","year":"2016"},{"key":"ref230","first-page":"3918","article-title":"Parallel WaveNet: Fast high-fidelity speech synthesis","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"van den Oord"},{"key":"ref231","first-page":"4797","article-title":"Conditional image generation with PixelCNN decoders","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"van den Oord"},{"key":"ref232","doi-asserted-by":"publisher","DOI":"10.1002\/9781119214403.ch5"},{"key":"ref233","first-page":"6309","article-title":"Neural discrete representation learning","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"van den Oord"},{"key":"ref234","first-page":"6000","article-title":"Attention is all you need","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Vaswani"},{"key":"ref235","doi-asserted-by":"publisher","DOI":"10.1090\/gsm\/058"},{"key":"ref236","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-71050-9"},{"key":"ref237","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00142"},{"key":"ref238","first-page":"30","article-title":"BERT has a mouth, and it must speak: BERT as a Markov random field language model","volume-title":"Proc. Workshop Methods Optimizing Evaluating Neural Lang. Gener.","author":"Wang"},{"key":"ref239","article-title":"Linformer: Self-attention with linear complexity","author":"Wang","year":"2020"},{"key":"ref240","article-title":"On solving minimax optimization locally: A follow-the-ridge approach","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Wang"},{"key":"ref241","article-title":"Learning to efficiently sample from diffusion probabilistic models","author":"Watson","year":"2021"},{"key":"ref242","first-page":"1545","article-title":"Unconstrained Monotonic neural networks","volume-title":"Proc. 33rd Conf. Neural Inf. Process. Syst.","author":"Wehenkel"},{"key":"ref243","first-page":"681","article-title":"Bayesian learning via stochastic gradient langevin dynamics","volume-title":"Proc. 28th Int. Conf. Mach. Learn.","author":"Welling"},{"key":"ref244","first-page":"5933","article-title":"Stochastic normalizing flows","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Wu"},{"key":"ref245","article-title":"VAEBM: A symbiosis between variational autoencoders and energy-based models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Xiao"},{"key":"ref246","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2879081"},{"key":"ref247","first-page":"2635","article-title":"A theory of generative ConvNet","volume-title":"Proc. 33rd Int. Conf. Mach. Learn.","author":"Xie"},{"key":"ref248","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3028042"},{"key":"ref249","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101552"},{"key":"ref250","first-page":"10957","article-title":"Training deep energy-based models with f-divergence minimization","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","author":"Yu"},{"key":"ref251","doi-asserted-by":"publisher","DOI":"10.5244\/C.30.87"},{"key":"ref252","article-title":"Approximation capabilities of neural ordinary differential equations","author":"Zhang","year":"2019"},{"key":"ref253","first-page":"7354","article-title":"Self-attention generative adversarial networks","author":"Zhang","year":"2019","journal-title":"Proc. Int. Conf. Mach. Learn."},{"key":"ref254","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.629"},{"key":"ref255","article-title":"Monge-Amp\u00e8re flow for generative modeling","author":"Zhang","year":"2018"},{"key":"ref256","article-title":"Energy-based generative adversarial network","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zhao"},{"key":"ref257","first-page":"4091","article-title":"Learning hierarchical features from deep generative models","volume-title":"Proc. 34th Int. Conf. Mach. Learn.","author":"Zhao"},{"key":"ref258","article-title":"Towards deeper understanding of variational autoencoding models","author":"Zhao","year":"2017"},{"key":"ref259","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33015885"},{"key":"ref260","first-page":"7559","article-title":"Differentiable augmentation for data-efficient GAN training","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Zhao"},{"key":"ref261","article-title":"Image augmentations for GAN training","author":"Zhao","year":"2020"},{"key":"ref262","article-title":"Improving the speed and quality of GAN by adversarial training","author":"Zhong","year":"2020"},{"key":"ref263","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.244"},{"key":"ref264","first-page":"7673","article-title":"Latent normalizing flows for discrete sequences","volume-title":"Proc. 36th Int. Conf. Mach. Learn.","author":"Ziegler"},{"key":"ref265","article-title":"Learning latent space energy-based prior model","volume-title":"Proc. Conf. Neural Inf. Process. Syst.","author":"Pang"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/34\/9910240\/09555209.pdf?arnumber=9555209","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,11]],"date-time":"2024-01-11T23:06:15Z","timestamp":1705014375000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9555209\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,11,1]]},"references-count":265,"journal-issue":{"issue":"11"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2021.3116668","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":[[2022,11,1]]}}}