{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,27]],"date-time":"2025-07-27T07:18:25Z","timestamp":1753600705664,"version":"3.28.0"},"reference-count":21,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,5,23]]},"DOI":"10.1109\/icassp43922.2022.9746778","type":"proceedings-article","created":{"date-parts":[[2022,4,27]],"date-time":"2022-04-27T15:50:34Z","timestamp":1651074634000},"page":"4243-4247","source":"Crossref","is-referenced-by-count":4,"title":["Neural Collapse in Deep Homogeneous Classifiers and The Role of Weight Decay"],"prefix":"10.1109","author":[{"given":"Akshay","family":"Rangamani","sequence":"first","affiliation":[{"name":"MIT,Center for Brains, Minds, and Machines"}]},{"given":"Andrzej","family":"Banburski-Fahey","sequence":"additional","affiliation":[{"name":"MIT,Center for Brains, Minds, and Machines"}]}],"member":"263","reference":[{"key":"ref10","first-page":"322","article-title":"Fine-grained analysis of optimization and generalization for overparameterized two-layer neural networks","author":"arora","year":"2019","journal-title":"International Conference on Machine Learning"},{"key":"ref11","article-title":"Evaluation of neural architectures trained with square loss vs cross-entropy in classification tasks","author":"hui","year":"2020","journal-title":"arXiv preprint arXiv 2006 04989"},{"journal-title":"Everything Old Is New Again A Fresh Look at Historical Approaches to Machine Learning","year":"2002","author":"rifkin","key":"ref12"},{"key":"ref13","article-title":"Layer-peeled model: Toward understanding well-trained deep neural networks","volume":"abs 2101 12699","author":"fang","year":"2021","journal-title":"CoRR"},{"key":"ref14","article-title":"Neural collapse with unconstrained features","volume":"abs 2011 11619","author":"mixon","year":"2020","journal-title":"CoRR"},{"key":"ref15","article-title":"Neural collapse under MSE loss: Proximity to and dynamics on the central path","volume":"abs 2106 2073","author":"han","year":"2021","journal-title":"CoRR"},{"article-title":"A geometric analysis of neural collapse with unconstrained features","year":"2021","author":"zhu","key":"ref16"},{"key":"ref17","article-title":"Neural collapse with cross-entropy loss","volume":"abs 2012 8465","author":"lu","year":"2020","journal-title":"CoRR"},{"key":"ref18","article-title":"On the emergence of tetrahedral symmetry in the final and penultimate layers of neural network classifiers","author":"wojtowytsch","year":"2020","journal-title":"2012 arXiv preprint arXiv"},{"key":"ref19","article-title":"Revealing the structure of deep neural networks via convex duality","author":"ergen","year":"2020","journal-title":"arXiv preprint arXiv 2002 01243"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1907369117"},{"key":"ref3","article-title":"Gradient descent maximizes the margin of homogeneous neural networks","volume":"abs 1906 5890","author":"lyu","year":"2019","journal-title":"CoRR"},{"key":"ref6","first-page":"1305","article-title":"Implicit bias of gradient descent for wide two-layer neural networks trained with the logistic loss","author":"chizat","year":"2020","journal-title":"Conference on Learning Theory"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-020-14663-9"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1806579115"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1002\/sta4.354"},{"key":"ref2","first-page":"2822","article-title":"The implicit bias of gradient descent on separable data","volume":"19","author":"soudry","year":"2018","journal-title":"The Journal of Machine Learning Research"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2015509117"},{"key":"ref9","article-title":"A dynamical central limit theorem for shallow neural networks","author":"chen","year":"2020","journal-title":"arXiv preprint arXiv 2008 06439"},{"key":"ref20","article-title":"Generalization in deep network classifiers trained with the square loss","author":"poggio","year":"2021","journal-title":"Center for Brains Minds & Machines (CBMM) Memo"},{"key":"ref21","article-title":"Dynamics and neural collapse in deep classifiers trained with the square loss","author":"banburski","year":"2021","journal-title":"Center for Brains Minds & Machines (CBMM) Memo"}],"event":{"name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2022,5,23]]},"location":"Singapore, Singapore","end":{"date-parts":[[2022,5,27]]}},"container-title":["ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9745891\/9746004\/09746778.pdf?arnumber=9746778","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,15]],"date-time":"2022-08-15T16:05:19Z","timestamp":1660579519000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9746778\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,23]]},"references-count":21,"URL":"https:\/\/doi.org\/10.1109\/icassp43922.2022.9746778","relation":{},"subject":[],"published":{"date-parts":[[2022,5,23]]}}}