Modern deep neural networks for classification usually jointly learn a backbone for
representation and a linear classifier to output the logit of each class. A recent study has …

One of the main challenges in feature learning using Deep Convolutional Neural Networks
(DCNNs) for large-scale face recognition is the design of appropriate loss functions that can …

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 …

In class-incremental learning, a learning agent faces a stream of data with the goal of
learning new classes while not forgetting previous ones. Neural networks are known to …

Learning compatible representations enables the interchangeable use of semantic features
as models are updated over time. This is particularly relevant in search and retrieval systems …

There is a recently discovered and intriguing phenomenon called Neural Collapse: at the
terminal phase of training a deep neural network for classification, the within-class …

Neural networks are widely used as a model for classification in a large variety of tasks.
Typically, a learnable transformation (ie, the classifier) is placed at the end of such models …

Making each modality in multi-modal data contribute is of vital importance to learning a
versatile multi-modal model. Existing methods, however, are often dominated by one or few …

Compatible features enable the direct comparison of old and new learned features allowing
to use them interchangeably over time. In visual search systems, this eliminates the need to …

In this article, we exploit Semi-Supervised Learning (SSL) to increase the amount of training
data to improve the performance of Fine-Grained Visual Categorization (FGVC). This …