{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T03:10:04Z","timestamp":1781665804390,"version":"3.54.5"},"reference-count":47,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2021,5,1]],"date-time":"2021-05-01T00:00:00Z","timestamp":1619827200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,5,1]],"date-time":"2021-05-01T00:00:00Z","timestamp":1619827200000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,5,1]],"date-time":"2021-05-01T00:00:00Z","timestamp":1619827200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,5,1]],"date-time":"2021-05-01T00:00:00Z","timestamp":1619827200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"U.S. National Science Foundation","doi-asserted-by":"publisher","award":["CCF-1801855"],"award-info":[{"award-number":["CCF-1801855"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"U.S. National Science Foundation","doi-asserted-by":"publisher","award":["CCF-1900145"],"award-info":[{"award-number":["CCF-1900145"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Inform. Theory"],"published-print":{"date-parts":[[2021,5]]},"DOI":"10.1109\/tit.2021.3053234","type":"journal-article","created":{"date-parts":[[2021,1,21]],"date-time":"2021-01-21T21:19:26Z","timestamp":1611263966000},"page":"3114-3129","source":"Crossref","is-referenced-by-count":14,"title":["Understanding Estimation and Generalization Error of Generative Adversarial Networks"],"prefix":"10.1109","volume":"67","author":[{"given":"Kaiyi","family":"Ji","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yingbin","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","article-title":"On generalization error bounds of noisy gradient methods for non-convex learning","author":"li","year":"2020","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref38","first-page":"2608","article-title":"Stagewise training accelerates convergence of testing error over SGD","author":"yuan","year":"2019","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref33","first-page":"3100","article-title":"Improved zeroth-order variance reduced algorithms and analysis for nonconvex optimization","author":"ji","year":"2019","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref32","first-page":"745","article-title":"Stability and generalization of learning algorithms that converge to global optima","author":"charles","year":"2018","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref31","article-title":"Generalization error bounds with probabilistic guarantee for SGD in nonconvex optimization","author":"zhou","year":"2018","journal-title":"arXiv 1802 06903"},{"key":"ref30","first-page":"1998","article-title":"Gradient diversity: A key ingredient for scalable distributed learning","author":"yin","year":"2018","journal-title":"Proc Int Conf Artif Intell Statist (AISTATS)"},{"key":"ref37","first-page":"1519","article-title":"Stochastic composite mirror descent: Optimal bounds with high probabilities","author":"lei","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref36","first-page":"1270","article-title":"High probability generalization bounds for uniformly stable algorithms with nearly optimal rate","volume":"99","author":"feldman","year":"2019","journal-title":"Proc Conf Learn Theory (COLT)"},{"key":"ref35","first-page":"9747","article-title":"Generalization bounds for uniformly stable algorithms","author":"feldman","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref34","article-title":"Online learning, stability, and stochastic gradient descent","author":"poggio","year":"2011","journal-title":"arXiv 1105 4701"},{"key":"ref10","article-title":"Improving generalization and stability of generative adversarial networks","author":"thanh-tung","year":"2019","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.278"},{"key":"ref11","first-page":"5551","article-title":"Approximation and convergence properties of generative adversarial learning","author":"liu","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref12","first-page":"10225","article-title":"Nonparametric density estimation under adversarial losses","author":"singh","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref13","article-title":"How well generative adversarial networks learn distributions","author":"liang","year":"2018","journal-title":"arXiv 1811 03179"},{"key":"ref14","first-page":"9089","article-title":"Nonparametric density estimation & convergence rates for GANs under Besov IPM losses","author":"uppal","year":"2019","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref15","first-page":"1","article-title":"Regularisation of neural networks by enforcing Lipschitz continuity","author":"gouk","year":"2020","journal-title":"Mach Learn"},{"key":"ref16","article-title":"Lipschitz regularized deep neural networks generalize and are adversarially robust","author":"finlay","year":"2018","journal-title":"arXiv 1808 09540"},{"key":"ref17","first-page":"9725","article-title":"Data-dependent sample complexity of deep neural networks via Lipschitz augmentation","author":"wei","year":"2019","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref18","article-title":"When will gradient methods converge to max-margin classifier under ReLU models?","author":"xu","year":"2018","journal-title":"arXiv 1806 04339"},{"key":"ref19","author":"anthony","year":"2009","journal-title":"Neural Network Learning Theoretical Foundations"},{"key":"ref28","first-page":"2815","article-title":"Data-dependent stability of stochastic gradient descent","author":"kuzborskij","year":"2018","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref4","article-title":"How well can generative adversarial networks learn densities: A nonparametric view","author":"liang","year":"2017","journal-title":"arXiv 1712 08244"},{"key":"ref27","first-page":"2635","article-title":"Learnability, stability and uniform convergence","volume":"11","author":"shalev-shwartz","year":"2010","journal-title":"J Mach Learn Res"},{"key":"ref3","article-title":"On the discrimination-generalization tradeoff in GANs","author":"zhang","year":"2018","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref6","first-page":"5585","article-title":"Gradient descent GAN optimization is locally stable","author":"nagarajan","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref29","first-page":"605","article-title":"Generalization bounds of sgld for non-convex learning: Two theoretical viewpoints","author":"mou","year":"2018","journal-title":"Proc Conf Learn Theory (COLT)"},{"key":"ref5","first-page":"1225","article-title":"Train faster, generalize better: Stability of stochastic gradient descent","author":"hardt","year":"2016","journal-title":"Proc 33rd Int Conf Mach Learn (ICML)"},{"key":"ref8","first-page":"258","article-title":"Training generative neural networks via maximum mean discrepancy optimization","author":"dziugaite","year":"2015","journal-title":"Proc 31st Conf Uncertainty Artif Intell (UAI)"},{"key":"ref7","article-title":"Unsupervised representation learning with deep convolutional generative adversarial networks","author":"radford","year":"2015","journal-title":"arXiv 1511 06434"},{"key":"ref2","first-page":"224","article-title":"Generalization and equilibrium in generative adversarial nets (GANs)","author":"arora","year":"2017","journal-title":"Proc 34th Int Conf Mach Learn (ICML)"},{"key":"ref9","article-title":"Understanding GANs via generalization analysis for disconnected support","author":"imaizumi","year":"2019","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref1","first-page":"2672","article-title":"Generative adversarial nets","author":"goodfellow","year":"2014","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref46","article-title":"Mode regularized generative adversarial networks","author":"che","year":"2017","journal-title":"Proc Int Conf Learn Represent (ICLR)"},{"key":"ref20","first-page":"6240","article-title":"Spectrally-normalized margin bounds for neural networks","author":"bartlett","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref45","first-page":"6626","article-title":"GANs trained by a two time-scale update rule converge to a local Nash equilibrium","author":"heusel","year":"2017","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref22","first-page":"1376","article-title":"Norm-based capacity control in neural networks","author":"neyshabur","year":"2015","journal-title":"Proc Conf Learn Theory (COLT)"},{"key":"ref47","author":"tsybakov","year":"2008","journal-title":"Introduction to Nonparametric Estimation"},{"key":"ref21","first-page":"297","article-title":"Size-independent sample complexity of neural networks","author":"golowich","year":"2018","journal-title":"Proc Conf Learn Theory (COLT)"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.244"},{"key":"ref24","first-page":"3845","article-title":"Minimax estimation of neural net distance","author":"ji","year":"2018","journal-title":"Proc Adv Neural Inf Process Syst (NeurIPS)"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.632"},{"key":"ref23","first-page":"3966","article-title":"Learning compact neural networks with regularization","author":"oymak","year":"2018","journal-title":"Proc Int Conf Mach Learn (ICML)"},{"key":"ref44","article-title":"An empirical study on evaluation metrics of generative adversarial networks","author":"xu","year":"2018","journal-title":"arXiv 1806 07755"},{"key":"ref26","first-page":"55","article-title":"Stability of randomized learning algorithms","volume":"6","author":"elisseeff","year":"2005","journal-title":"J Mach Learn Res"},{"key":"ref43","first-page":"901","article-title":"Weight normalization: A simple reparameterization to accelerate training of deep neural networks","author":"salimans","year":"2016","journal-title":"Proc Adv Neural Inf Process Syst (NIPS)"},{"key":"ref25","first-page":"499","article-title":"Stability and generalization","volume":"2","author":"bousquet","year":"2002","journal-title":"J Mach Learn Res"}],"container-title":["IEEE Transactions on Information Theory"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/18\/9409872\/9330788-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/18\/9409872\/09330788.pdf?arnumber=9330788","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T14:54:14Z","timestamp":1652194454000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9330788\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,5]]},"references-count":47,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tit.2021.3053234","relation":{},"ISSN":["0018-9448","1557-9654"],"issn-type":[{"value":"0018-9448","type":"print"},{"value":"1557-9654","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,5]]}}}