{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T16:29:18Z","timestamp":1781368158485,"version":"3.54.1"},"reference-count":60,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,5,1]],"date-time":"2025-05-01T00:00:00Z","timestamp":1746057600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation (NSF) under CAREER Award","doi-asserted-by":"publisher","award":["CCF-2236829"],"award-info":[{"award-number":["CCF-2236829"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"name":"U.S. Army Research Office Early Career Award","award":["W911NF-21-1-0242"],"award-info":[{"award-number":["W911NF-21-1-0242"]}]},{"DOI":"10.13039\/100000006","name":"Office of Naval Research","doi-asserted-by":"publisher","award":["N00014-24-1-2164"],"award-info":[{"award-number":["N00014-24-1-2164"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Inform. Theory"],"published-print":{"date-parts":[[2025,5]]},"DOI":"10.1109\/tit.2025.3545564","type":"journal-article","created":{"date-parts":[[2025,2,25]],"date-time":"2025-02-25T18:47:12Z","timestamp":1740509232000},"page":"3854-3870","source":"Crossref","is-referenced-by-count":4,"title":["The Convex Landscape of Neural Networks: Characterizing Global Optima and Stationary Points via Lasso Models"],"prefix":"10.1109","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4806-0224","authenticated-orcid":false,"given":"Tolga","family":"Ergen","sequence":"first","affiliation":[{"name":"LG AI Research, Ann Arbor, MI, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0870-9992","authenticated-orcid":false,"given":"Mert","family":"Pilanci","sequence":"additional","affiliation":[{"name":"Electrical Engineering Department, Stanford University, Stanford, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"123","article-title":"Convex neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"18","author":"Bengio"},{"issue":"1","key":"ref2","first-page":"629","article-title":"Breaking the curse of dimensionality with convex neural networks","volume":"18","author":"Bach","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2022.3163341"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2004.834793"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2005.858979"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2006.871582"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1137\/S1064827596304010"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1214\/009053606000001523"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-9868.2005.00532.x"},{"key":"ref10","first-page":"7695","article-title":"Neural networks are convex regularizers: Exact polynomial-time convex optimization formulations for two-layer networks","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Pilanci"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1090\/s0002-9947-1975-0367131-6"},{"key":"ref12","article-title":"Understanding deep neural networks with rectified linear units","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Arora"},{"key":"ref13","first-page":"1004","article-title":"Reliably learning the relu in polynomial time","volume-title":"Proc. Conf. Learn. Theory","volume":"65","author":"Goel"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1090\/pcms\/013\/08"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/PGEC.1965.264137"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1016\/j.disopt.2023.100795"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1016\/j.disopt.2020.100620"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511804441"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1137\/0215024"},{"key":"ref20","volume-title":"Convex Polytopes","volume":"16","author":"Gr\u00fcnbaum","year":"1967"},{"key":"ref21","first-page":"26","article-title":"An elementary introduction to modern convex geometry","volume":"31","author":"Ball","year":"1997","journal-title":"Flavors of Geometry"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1137\/070697835"},{"key":"ref24","article-title":"Matrix rank minimization with applications","author":"Fazel","year":"2002"},{"key":"ref25","first-page":"1","article-title":"Implicit bias of gradient descent on linear convolutional networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Gunasekar"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1111\/j.2517-6161.1996.tb02080.x"},{"key":"ref27","first-page":"8571","article-title":"Neural tangent kernel: Convergence and generalization in neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Jacot"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/lsp.2020.3027517"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1137\/21m1418642"},{"key":"ref30","article-title":"How do infinite width bounded norm networks look in function space?","author":"Savarese","year":"2019","journal-title":"arXiv:1902.05040"},{"key":"ref31","article-title":"On lazy training in differentiable programming","volume-title":"arXiv:1812.07956","author":"Chizat","year":"2020"},{"key":"ref32","first-page":"1305","article-title":"Implicit bias of gradient descent for wide two-layer neural networks trained with the logistic loss","volume-title":"Proc. Conf. Learn. Theory","author":"Chizat"},{"key":"ref33","first-page":"3635","article-title":"Kernel and rich regimes in overparametrized models","volume-title":"Proc. Thirty 3rd Conf. Learn. Theory","author":"Woodworth"},{"key":"ref34","first-page":"1","article-title":"Implicit bias in deep linear classification: Initialization scale vs training accuracy","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Moroshko"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1016\/0166-218X(95)00026-N"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1201\/9781420035315.ch24"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1137\/15M1027930"},{"key":"ref38","first-page":"4024","article-title":"Convex geometry of two-layer ReLU networks: Implicit autoencoding and interpretable models","volume-title":"Proc. Int. Conf. Artif. Intell. Statist","volume":"108","author":"Ergen"},{"issue":"212","key":"ref39","first-page":"1","article-title":"Convex geometry and duality of over-parameterized neural networks","volume":"22","author":"Ergen","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"ref40","first-page":"1","article-title":"Vector-output ReLU neural network problems are copositive programs: Convex analysis of two layer networks and polynomial-time algorithms","volume-title":"Proc. Int. Conf. Learn. Represent","author":"Sahiner"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413662"},{"key":"ref42","first-page":"1","article-title":"Implicit regularization in deep matrix factorization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Arora"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ita.2018.8503198"},{"key":"ref44","first-page":"21174","article-title":"Implicit regularization in deep learning may not be explainable by norms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Razin"},{"key":"ref45","author":"Grant","year":"2014","journal-title":"CVX: MATLAB Software for Disciplined Convex Programming"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1161\/01.CTR.101.23.e215"},{"key":"ref47","author":"Dua","year":"2017","journal-title":"UCI Machine Learning Repository"},{"key":"ref48","first-page":"1","article-title":"Harnessing the power of infinitely wide deep nets on small-data tasks","volume-title":"Proc. Int. Conf. Learn. Represent","author":"Arora"},{"key":"ref49","author":"Krizhevsky","year":"2014","journal-title":"The CIFAR-10 Dataset"},{"key":"ref50","article-title":"Fashion-MNIST: A novel image dataset for benchmarking machine learning algorithms","volume-title":"arXiv:1708.07747","author":"Xiao","year":"2017"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-35289-8_30"},{"key":"ref52","first-page":"215","article-title":"An analysis of single-layer networks in unsupervised feature learning","volume-title":"Proc. 14th Int. Conf. Artif. Intell. Statist.","author":"Coates"},{"key":"ref53","article-title":"The hidden convex optimization landscape of two-layer ReLU neural networks: An exact characterization of the optimal solutions","volume-title":"arXiv:2006.05900","author":"Wang","year":"2022"},{"key":"ref54","first-page":"1","article-title":"Implicit convex regularizers of CNN architectures: Convex optimization of two- and three-layer networks in polynomial time","volume-title":"Proc. Int. Conf. Learn. Represent","author":"Ergen"},{"key":"ref55","first-page":"3004","article-title":"Revealing the structure of deep neural networks via convex duality","volume-title":"Proc. 38th Int. Conf. Mach. Learn","author":"Ergen"},{"key":"ref56","first-page":"1","article-title":"Hidden convexity of Wasserstein GANs: Interpretable generative models with closed-form solutions","volume-title":"Proc. Int. Conf. Learn. Represent","author":"Sahiner"},{"key":"ref57","first-page":"2993","article-title":"Global optimality beyond two layers: Training deep ReLU networks via convex programs","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Ergen"},{"key":"ref58","first-page":"19050","article-title":"Unraveling attention via convex duality: Analysis and interpretations of vision transformers","volume-title":"Proc. 39th Int. Conf. Mach. Learn","volume":"162","author":"Sahiner"},{"key":"ref59","article-title":"Convexifying transformers: Improving optimization and understanding of transformer networks","author":"Ergen","year":"2022","journal-title":"arXiv:2211.11052"},{"key":"ref60","first-page":"1","article-title":"Demystifying batch normalization in ReLU networks: Equivalent convex optimization models and implicit regularization","volume-title":"Proc. Int. Conf. Learn. Represent","author":"Ergen"}],"container-title":["IEEE Transactions on Information Theory"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/18\/10975818\/10902486-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/18\/10975818\/10902486.pdf?arnumber=10902486","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,25]],"date-time":"2025-04-25T17:40:01Z","timestamp":1745602801000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10902486\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5]]},"references-count":60,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tit.2025.3545564","relation":{},"ISSN":["0018-9448","1557-9654"],"issn-type":[{"value":"0018-9448","type":"print"},{"value":"1557-9654","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5]]}}}