{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T12:07:04Z","timestamp":1784203624870,"version":"3.55.0"},"reference-count":44,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Neurocomputing"],"published-print":{"date-parts":[[2026,10]]},"DOI":"10.1016\/j.neucom.2026.134176","type":"journal-article","created":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T16:28:33Z","timestamp":1780417713000},"page":"134176","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Multi-time dynamics in neural network optimization: A unified framework bridging game theory and optimal control"],"prefix":"10.1016","volume":"696","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3663-836X","authenticated-orcid":false,"given":"Massimiliano","family":"Ferrara","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"issue":"3","key":"10.1016\/j.neucom.2026.134176_bib0005","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1145\/3446776","article-title":"Understanding deep learning (still) requires rethinking generalization","volume":"64","author":"Zhang","year":"2021","journal-title":"Commun. ACM"},{"key":"10.1016\/j.neucom.2026.134176_bib0010","series-title":"International Conference on Learning Representations","article-title":"Exact solutions to the nonlinear dynamics of learning in deep linear neural networks","author":"Saxe","year":"2014"},{"key":"10.1016\/j.neucom.2026.134176_bib0015","series-title":"International Conference on Machine Learning, PMLR","first-page":"244","article-title":"On the optimization of deep networks: implicit acceleration by overparameterization","author":"Arora","year":"2018"},{"key":"10.1016\/j.neucom.2026.134176_bib0020","article-title":"How transferable are features in deep neural networks?","volume":"27","author":"Yosinski","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.neucom.2026.134176_bib0025","series-title":"Convex Functions and Optimization Methods on Riemannian Manifolds","author":"Udri\u015fte","year":"1994"},{"key":"10.1016\/j.neucom.2026.134176_bib0030","series-title":"Geometric Dynamics","author":"Udri\u015fte","year":"2000"},{"key":"10.1016\/j.neucom.2026.134176_bib0035","series-title":"International Conference on Machine Learning, PMLR","first-page":"1675","article-title":"Gradient descent finds global minima of deep neural networks","author":"Du","year":"2019"},{"key":"10.1016\/j.neucom.2026.134176_bib0040","series-title":"International Conference on Machine Learning, PMLR","first-page":"242","article-title":"A convergence theory for deep learning via over-parameterization","author":"Allen-Zhu","year":"2019"},{"key":"10.1016\/j.neucom.2026.134176_bib0045","article-title":"Generative adversarial nets","volume":"27","author":"Goodfellow","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.neucom.2026.134176_bib0050","series-title":"International Conference on Machine Learning, PMLR","first-page":"3481","article-title":"Which training methods for GANs do actually converge?","author":"Mescheder","year":"2018"},{"key":"10.1016\/j.neucom.2026.134176_bib0055","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1007\/978-3-030-60990-0_12","article-title":"Multi-agent reinforcement learning: a selective overview of theories and algorithms","author":"Zhang","year":"2021","journal-title":"Handbook of Reinforcement Learning and Control"},{"issue":"2","key":"10.1016\/j.neucom.2026.134176_bib0060","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1162\/089976698300017746","article-title":"Natural gradient works efficiently in learning","volume":"10","author":"Amari","year":"1998","journal-title":"Neural Comput."},{"issue":"146","key":"10.1016\/j.neucom.2026.134176_bib0065","first-page":"1","article-title":"New insights and perspectives on the natural gradient method","volume":"21","author":"Martens","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"10.1016\/j.neucom.2026.134176_bib0070","series-title":"Geometric Numerical Integration: Structure-Preserving Algorithms for Ordinary Differential Equations","author":"Hairer","year":"2006"},{"key":"10.1016\/j.neucom.2026.134176_bib0075","series-title":"Optimization Algorithms on Matrix Manifolds","author":"Absil","year":"2008"},{"issue":"2","key":"10.1016\/j.neucom.2026.134176_bib0080","first-page":"138","article-title":"Multitime optimal control with area integral costs on boundary","volume":"15","author":"Udri\u015fte","year":"2010","journal-title":"Balkan Journal of Geometry and Its Applications"},{"issue":"5","key":"10.1016\/j.neucom.2026.134176_bib0085","first-page":"401","article-title":"On a class of optimal control problems governed by multitime Hamilton-Jacobi PDES","volume":"11","author":"Pitea","year":"2012","journal-title":"WSEAS Transactions on Mathematics"},{"issue":"3","key":"10.1016\/j.neucom.2026.134176_bib0090","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1215\/S0012-7094-41-00838-4","article-title":"A generalization of brouwer\u2019s fixed point theorem","volume":"8","author":"Kakutani","year":"1941","journal-title":"Duke Math. J."},{"key":"10.1016\/j.neucom.2026.134176_bib0095","series-title":"The Mathematical Theory of Optimal Processes","author":"Pontryagin","year":"1962"},{"key":"10.1016\/j.neucom.2026.134176_bib0100","series-title":"International Conference on Learning Representations","article-title":"Large batch optimization for deep learning: training BERT in 76min","author":"You","year":"2019"},{"key":"10.1016\/j.neucom.2026.134176_bib0105","series-title":"Technical report","article-title":"Learning Multiple Layers of Features From Tiny Images","author":"Krizhevsky","year":"2009"},{"key":"10.1016\/j.neucom.2026.134176_bib0110","author":"Xiao"},{"issue":"3","key":"10.1016\/j.neucom.2026.134176_bib0115","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","article-title":"ImageNet large scale visual recognition challenge","volume":"115","author":"Russakovsky","year":"2015","journal-title":"Int. J. Comput. Vis."},{"key":"10.1016\/j.neucom.2026.134176_bib0120","series-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.1016\/j.neucom.2026.134176_bib0125","author":"Simonyan"},{"key":"10.1016\/j.neucom.2026.134176_bib0130","doi-asserted-by":"crossref","first-page":"400","DOI":"10.1214\/aoms\/1177729586","article-title":"A stochastic approximation method","volume":"3","author":"Robbins","year":"1951","journal-title":"The Annals of Mathematical Statistics 22"},{"key":"10.1016\/j.neucom.2026.134176_bib0135","author":"Kingma"},{"key":"10.1016\/j.neucom.2026.134176_bib0140","first-page":"2121","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"Duchi","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"10.1016\/j.neucom.2026.134176_bib0145","series-title":"International Conference on Machine Learning, PMLR","first-page":"3519","article-title":"Similarity of neural network representations revisited","author":"Kornblith","year":"2019"},{"key":"10.1016\/j.neucom.2026.134176_bib0150","article-title":"SVCCA: singular vector canonical correlation analysis for deep learning dynamics and interpretability","volume":"30","author":"Raghu","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"issue":"40","key":"10.1016\/j.neucom.2026.134176_bib0155","doi-asserted-by":"crossref","first-page":"24652","DOI":"10.1073\/pnas.2015509117","article-title":"Prevalence of neural collapse during the terminal phase of deep learning training","volume":"117","author":"Papyan","year":"2020","journal-title":"Proc. Natl. Acad. Sci."},{"key":"10.1016\/j.neucom.2026.134176_bib0160","series-title":"International Conference on Learning Representations","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2019"},{"key":"10.1016\/j.neucom.2026.134176_bib0165","article-title":"Symbolic discovery of optimization algorithms","volume":"36","author":"Chen","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.neucom.2026.134176_bib0170","series-title":"International Conference on Machine Learning, PMLR","first-page":"1842","article-title":"Shampoo: preconditioned stochastic tensor optimization","author":"Gupta","year":"2018"},{"key":"10.1016\/j.neucom.2026.134176_bib0175","series-title":"International Conference on Learning Representations","article-title":"SOAP: improving and stabilizing shampoo using Adam","author":"Vyas","year":"2025"},{"key":"10.1016\/j.neucom.2026.134176_bib0180","series-title":"Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics","first-page":"328","article-title":"Universal language model fine-tuning for text classification","author":"Howard","year":"2018"},{"key":"10.1016\/j.neucom.2026.134176_bib0185","article-title":"Neural ordinary differential equations","volume":"31","author":"Chen","year":"2018","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"10.1016\/j.neucom.2026.134176_bib0190","series-title":"Mathematical Surveys and Monographs","article-title":"A tour of subriemannian geometries, their geodesics and applications","volume":"vol. 91","author":"Montgomery","year":"2002"},{"key":"10.1016\/j.neucom.2026.134176_bib0195","series-title":"Infinite Dimensional Analysis: A Hitchhiker\u2019s Guide","author":"Aliprantis","year":"2006"},{"key":"10.1016\/j.neucom.2026.134176_bib0200","doi-asserted-by":"crossref","DOI":"10.1016\/j.engappai.2025.113013","article-title":"Adaptive hierarchical control of quadcopters via safe reinforcement learning from human demonstration","volume":"163","author":"Tan","year":"2026","journal-title":"Eng. Appl. Artif. Intell."},{"key":"10.1016\/j.neucom.2026.134176_bib0205","doi-asserted-by":"crossref","first-page":"19568","DOI":"10.1109\/TASE.2025.3596912","article-title":"Hierarchical safe reinforcement learning control for leader-follower systems with prescribed performance","volume":"22","author":"Tan","year":"2025","journal-title":"IEEE Trans. Autom. Sci. Eng."},{"issue":"2","key":"10.1016\/j.neucom.2026.134176_bib0210","doi-asserted-by":"crossref","DOI":"10.1109\/TIE.2025.3603074","article-title":"Fixed-time stochastic learning from human-UAV interaction with state-input constraints","volume":"73","author":"Tan","year":"2026","journal-title":"IEEE Trans. Ind. Electron."},{"issue":"2","key":"10.1016\/j.neucom.2026.134176_bib0215","first-page":"282","article-title":"Multi-time evolution models in deep learning dynamics","volume":"47","author":"Ferrara","year":"2025","journal-title":"Journal of Indian Academy of Mathematics"},{"key":"10.1016\/j.neucom.2026.134176_bib0220","doi-asserted-by":"crossref","DOI":"10.1007\/s10957-026-02958-8","article-title":"Geometric-entropic optimization: integrating optimal transport with riemannian gradient methods for neural network training","author":"Ferrara","year":"2026","journal-title":"J. Optim. Theory Appl."}],"container-title":["Neurocomputing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0925231226015742?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0925231226015742?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T11:43:21Z","timestamp":1784202201000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0925231226015742"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10]]},"references-count":44,"alternative-id":["S0925231226015742"],"URL":"https:\/\/doi.org\/10.1016\/j.neucom.2026.134176","relation":{},"ISSN":["0925-2312"],"issn-type":[{"value":"0925-2312","type":"print"}],"subject":[],"published":{"date-parts":[[2026,10]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Multi-time dynamics in neural network optimization: A unified framework bridging game theory and optimal control","name":"articletitle","label":"Article Title"},{"value":"Neurocomputing","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.neucom.2026.134176","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 The Author(s). Published by Elsevier B.V.","name":"copyright","label":"Copyright"}],"article-number":"134176"}}