DropCluster: A structured dropout for convolutional networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Liyan, Mordohai, Philippos, Aydore, Sergul
Format: Preprint
Published: 2020
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915322336903168
author Chen, Liyan
Mordohai, Philippos
Aydore, Sergul
author_facet Chen, Liyan
Mordohai, Philippos
Aydore, Sergul
contents Dropout as a common regularizer to prevent overfitting in deep neural networks has been less effective in convolutional layers than in fully connected layers. This is because Dropout drops features randomly, without considering local structure. When features are spatially correlated, as in the case of convolutional layers, information from the dropped features can still propagate to subsequent layers via neighboring features. To address this problem, structured forms of Dropout have been proposed. A drawback of these methods is that they do not adapt to the data. In this work, we leverage the structure in the outputs of convolutional layers and introduce a novel structured regularization method named DropCluster. Our approach clusters features in convolutional layers, and drops the resulting clusters randomly during training iterations. Experiments on CIFAR-10/100, SVHN, and APPA-REAL datasets demonstrate that our approach is effective and controls overfitting better than other approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2002_02997
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle DropCluster: A structured dropout for convolutional networks
Chen, Liyan
Mordohai, Philippos
Aydore, Sergul
Machine Learning
Computer Vision and Pattern Recognition
Dropout as a common regularizer to prevent overfitting in deep neural networks has been less effective in convolutional layers than in fully connected layers. This is because Dropout drops features randomly, without considering local structure. When features are spatially correlated, as in the case of convolutional layers, information from the dropped features can still propagate to subsequent layers via neighboring features. To address this problem, structured forms of Dropout have been proposed. A drawback of these methods is that they do not adapt to the data. In this work, we leverage the structure in the outputs of convolutional layers and introduce a novel structured regularization method named DropCluster. Our approach clusters features in convolutional layers, and drops the resulting clusters randomly during training iterations. Experiments on CIFAR-10/100, SVHN, and APPA-REAL datasets demonstrate that our approach is effective and controls overfitting better than other approaches.
title DropCluster: A structured dropout for convolutional networks
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2002.02997