Please use this identifier to cite or link to this item:
Type: Artigo
Title: Deep neural networks under stress
Author: Carvalho, Micael
Cord, Matthieu
Avila, Sandra
Thome, Nicolas
Valle, Eduardo
Abstract: In recent years, deep architectures have been used for transfer learning with state-of-the-art performance in many datasets. The properties of their features remain, however, largely unstudied under the transfer perspective. In this work, we present an extensive analysis of the resiliency of feature vectors extracted from deep models, with special focus on the trade-off between performance and compression rate. By introducing perturbations to image descriptions extracted from a deep convolutional neural network, we change their precision and number of dimensions, measuring how it affects the final score. We show that deep features are more robust to these disturbances when compared to classical approaches, achieving a compression rate of 98.4%, while losing only 0.88% of their original score for Pascal VOC 2007
Subject: Aprendizado profundo
Country: Estados Unidos
Editor: Institute of Electrical and Electronics Engineers
Rights: Fechado
Identifier DOI: 10.1109/ICIP.2016.7533200
Date Issue: 2016
Appears in Collections:FEEC - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
000390782004083.pdf1.11 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.