Neural collapse: A review on modelling principles and generalization
V Kothapalli - arXiv preprint arXiv:2206.04041, 2022 - arxiv.org
Deep classifier neural networks enter the terminal phase of training (TPT) when training
error reaches zero and tend to exhibit intriguing Neural Collapse (NC) properties. Neural …
error reaches zero and tend to exhibit intriguing Neural Collapse (NC) properties. Neural …
On the optimization landscape of neural collapse under mse loss: Global optimality with unconstrained features
When training deep neural networks for classification tasks, an intriguing empirical
phenomenon has been widely observed in the last-layer classifiers and features, where (i) …
phenomenon has been widely observed in the last-layer classifiers and features, where (i) …
Neural collapse under mse loss: Proximity to and dynamics on the central path
The recently discovered Neural Collapse (NC) phenomenon occurs pervasively in today's
deep net training paradigm of driving cross-entropy (CE) loss towards zero. During NC, last …
deep net training paradigm of driving cross-entropy (CE) loss towards zero. During NC, last …
Evaluation of neural architectures trained with square loss vs cross-entropy in classification tasks
Modern neural architectures for classification tasks are trained using the cross-entropy loss,
which is widely believed to be empirically superior to the square loss. In this work we …
which is widely believed to be empirically superior to the square loss. In this work we …
Human activity recognition through recurrent neural networks for human–robot interaction in agriculture
The present study deals with human awareness, which is a very important aspect of human–
robot interaction. This feature is particularly essential in agricultural environments, owing to …
robot interaction. This feature is particularly essential in agricultural environments, owing to …
Benign overfitting in multiclass classification: All roads lead to interpolation
K Wang, V Muthukumar… - Advances in Neural …, 2021 - proceedings.neurips.cc
The growing literature on" benign overfitting" in overparameterized models has been mostly
restricted to regression or binary classification settings; however, most success stories of …
restricted to regression or binary classification settings; however, most success stories of …
Microplastic pollution monitoring with holographic classification and deep learning
The observation and detection of the microplastic pollutants generated by industrial
manufacturing require the use of precise optical systems. Digital holography is well suited …
manufacturing require the use of precise optical systems. Digital holography is well suited …
Fault identification for photovoltaic systems using a multi-output deep learning approach
Fault classification and localization are imperative to maintaining an efficient photovoltaic
(PV) system. Due to the environmental factors that PV systems function in, they can be prone …
(PV) system. Due to the environmental factors that PV systems function in, they can be prone …
Inducing neural collapse in deep long-tailed learning
Although deep neural networks achieve tremendous success on various classification tasks,
the generalization ability drops sheer when training datasets exhibit long-tailed distributions …
the generalization ability drops sheer when training datasets exhibit long-tailed distributions …
Theoretical insights into multiclass classification: A high-dimensional asymptotic view
C Thrampoulidis, S Oymak… - Advances in Neural …, 2020 - proceedings.neurips.cc
Contemporary machine learning applications often involve classification tasks with many
classes. Despite their extensive use, a precise understanding of the statistical properties and …
classes. Despite their extensive use, a precise understanding of the statistical properties and …