Conference item icon

Conference item

Salient deconvolutional networks

Abstract:

Deconvolution is a popular method for visualizing deep convolutional neural networks; however, due to their heuristic nature, the meaning of deconvolutional visualizations is not entirely clear. In this paper, we introduce a family of reversed networks that generalizes and relates deconvolution, backpropagation and network saliency. We use this construction to thoroughly investigate and compare these methods in terms of quality and meaning of the produced images, and of what architectural cho...

Expand abstract
Publication status:
Published
Peer review status:
Peer reviewed
Version:
Accepted manuscript

Actions


Access Document


Files:
Publisher copy:
10.1007/978-3-319-46466-4_8

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS Division
Department:
Engineering Science
Oxford college:
New College; New College; New College; New College
Role:
Author
Publisher:
Springer Publisher's website
Publication date:
2016-09-17
Acceptance date:
2016-07-11
DOI:
EISSN:
1611-3349
ISSN:
0302-9743
Pubs id:
pubs:655292
URN:
uri:4a0a8851-1a3b-4af1-9fc7-0d796f9e09f5
UUID:
uuid:4a0a8851-1a3b-4af1-9fc7-0d796f9e09f5
Local pid:
pubs:655292
ISBN:
978-3-319-46465-7

Terms of use


Metrics


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP