Revisiting Orthogonality Regularization: A Study for Convolutional Neural Networks in Image Classification

Recent research in deep Convolutional Neural Networks(CNN) faces the challenges of vanishing/exploding gradient issues, training instability, and feature redundancy. Orthogonality Regularization(OR), which introduces a penalty function considering the orthogonality of neural networks, could be a rem...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Taehyeon Kim, Se-Young Yun
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	IEEE 2022-01-01
Schriftenreihe:	IEEE Access
Schlagworte:	Deep neural network (DNN) kernel orthogonality regularization convolutional neural network (CNN) regularization image classification
Online-Zugang:	https://ieeexplore.ieee.org/document/9804718/
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!