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) …
From lazy to rich to exclusive task representations in neural networks and neural codes
Neural circuits—both in the brain and in “artificial” neural network models—learn to solve a
remarkable variety of tasks, and there is a great current opportunity to use neural networks …
remarkable variety of tasks, and there is a great current opportunity to use neural networks …
Are all losses created equal: A neural collapse perspective
While cross entropy (CE) is the most commonly used loss function to train deep neural
networks for classification tasks, many alternative losses have been developed to obtain …
networks for classification tasks, many alternative losses have been developed to obtain …
Imbalance trouble: Revisiting neural-collapse geometry
C Thrampoulidis, GR Kini… - Advances in Neural …, 2022 - proceedings.neurips.cc
Neural Collapse refers to the remarkable structural properties characterizing the geometry of
class embeddings and classifier weights, found by deep nets when trained beyond zero …
class embeddings and classifier weights, found by deep nets when trained beyond zero …
ReduNet: A white-box deep network from the principle of maximizing rate reduction
This work attempts to provide a plausible theoretical framework that aims to interpret modern
deep (convolutional) networks from the principles of data compression and discriminative …
deep (convolutional) networks from the principles of data compression and discriminative …
Improving self-supervised learning by characterizing idealized representations
Despite the empirical successes of self-supervised learning (SSL) methods, it is unclear
what characteristics of their representations lead to high downstream accuracies. In this …
what characteristics of their representations lead to high downstream accuracies. In this …
Neural collapse with normalized features: A geometric analysis over the riemannian manifold
When training overparameterized deep networks for classification tasks, it has been widely
observed that the learned features exhibit a so-called" neural collapse'" phenomenon. More …
observed that the learned features exhibit a so-called" neural collapse'" phenomenon. More …
Sharpness-aware minimization leads to low-rank features
M Andriushchenko, D Bahri… - Advances in Neural …, 2023 - proceedings.neurips.cc
Sharpness-aware minimization (SAM) is a recently proposed method that minimizes the
sharpness of the training loss of a neural network. While its generalization improvement is …
sharpness of the training loss of a neural network. While its generalization improvement is …
Deep neural collapse is provably optimal for the deep unconstrained features model
P Súkeník, M Mondelli… - Advances in Neural …, 2024 - proceedings.neurips.cc
Neural collapse (NC) refers to the surprising structure of the last layer of deep neural
networks in the terminal phase of gradient descent training. Recently, an increasing amount …
networks in the terminal phase of gradient descent training. Recently, an increasing amount …