Average gradient outer product as a mechanism for deep neural collapse

D Beaglehole, P Súkeník, M Mondelli… - arXiv preprint arXiv …, 2024 - arxiv.org
Deep Neural Collapse (DNC) refers to the surprisingly rigid structure of the data
representations in the final layers of Deep Neural Networks (DNNs). Though the …

Compressed Event Sensing (CES) Volumes for Event Cameras

S Lin, Y Ma, J Chen, B Wen - International Journal of Computer Vision, 2024 - Springer
Deep learning has made significant progress in event-driven applications. But to match
standard vision networks, most approaches rely on aggregating events into grid-like …

Theoretical characterisation of the Gauss-Newton conditioning in Neural Networks

J Zhao, SP Singh, A Lucchi - arXiv preprint arXiv:2411.02139, 2024 - arxiv.org
The Gauss-Newton (GN) matrix plays an important role in machine learning, most evident in
its use as a preconditioning matrix for a wide family of popular adaptive methods to speed …

On the Predictability of Fine-Grained Cellular Network Throughput Using Machine Learning Models

O Basit, P Dinh, I Khan, ZJ Kong, YC Hu… - 2024 IEEE 21st …, 2024 - ieeexplore.ieee.org
Networking research has witnessed a renaissance from exploring the seemingly unlimited
predictive power of machine learning (ML) models. One such promising direction is …

On the feasibility of using an encoder-only model for anomaly detection: the BERTAD approach

JF Chaves-Tibaduiza, AI Becerra-Muñoz… - 2024 IEEE …, 2024 - ieeexplore.ieee.org
The Internet of Things has provided the necessary infrastructure for connected devices to
collaborate and share information intelligently. Detecting anomalies in the traffic generated …

SynapSign: an Advanced Machine Learning Framework for American Sign Language Recognition Utilizing a Novel Landmark-Based Dataset

ÖÖ Karadağ, İ Balikci, OO KaradaĞ… - 2024 Innovations in …, 2024 - ieeexplore.ieee.org
Hearing loss is a common condition affecting a significant proportion of the world's
population, creating barriers to effective communication. Sign language, particularly …

Mechanism of clean-priority learning in early stopped neural networks of infinite width

C Liu, A Abedsoltan, M Belkin - openreview.net
When random label noise is added to a training dataset, the prediction error of a neural
network on a label-noise-free test dataset initially improves during early training but …