Conference proceeding
Convolutional neural network weights regularization via orthogonalization


Authors/Editors

Research Areas
Currently no objects available

Publication Details
Author list: Gayer A
Publisher: SPIE Digital Library
Publication year: 2020
Title of series: Proceedings of SPIE Volume 11433
Number in series: 2
Volume number: 11433
Number of pages: 1040
ISBN: 9781510636439

Abstract

Regularization methods play an important role in artificial neural networks training, improving generalization performance and preventing them from overfitting. In this paper, we introduce a new regularization method, based on the orthogonalization of convolutional layer filters. Proposed method is easy to implement and it has plug-and-play compatibility with modern training approaches, without any changes or adaptations on their part. Experiments with MNIST and CIFAR10 datasets showed that the effectiveness of the suggested method depends on number of filters in the layer, and maximum increase in quality is achieved for architectures with small number of parameters, which is important for training fast and lightweight neural networks


Keywords
Currently no objects available

Documents
Currently no objects available

Last updated on 2021-14-04 at 01:04