Reference : Evolutionary Algorithms for Convolutional Neural Network Visualisation
Scientific congresses, symposiums and conference proceedings : Paper published in a book
Engineering, computing & technology : Computer science
http://hdl.handle.net/10993/40854
Evolutionary Algorithms for Convolutional Neural Network Visualisation
English
Bernard, Nicolas mailto [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Computer Science and Communications Research Unit (CSC)]
Leprévost, Franck mailto [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Computer Science and Communications Research Unit (CSC)]
2018
High Performance Computing -- 5th Latin American Conference, CARLA 2018, Piedecuesta, Colombia
Meneses, Esteban
Castro, Harold
Barrios Hernández, Carlos Jaime
Ramos-Pollan, Raúl
Springer
Communications in Computer and Information Science
Yes
978-3-030-16204-7
CARLA 2018 - 5th Latin America High Performance Computing Conference
from 23-09-2018 to 28-09-2018
Piedecuesta
Colombia
[en] Deep Learning is based on deep neural networks trained over huge sets of examples.
It enabled computers to compete with ---~or even outperform~--- humans at many tasks, from playing Go to driving vehicules.

Still, it remains hard to understand how these networks actually operate.
While an observer sees any individual local behaviour, he gets little insight about their global decision-making process.

However, there is a class of neural networks widely used for image processing, convolutional networks, where each layer contains features working in parallel.
By their structure, these features keep some spatial information across a network's layers.
Visualisation of this spatial information at different locations in a network, notably on input data that maximise the activation of a given feature, can give insights on the way the model works.

This paper investigates the use of Evolutionary Algorithms to evolve such input images that maximise feature activation.
Compared with some pre-existing approaches, ours seems currently computationally heavier but with a wider applicability.
University of Luxembourg: High Performance Computing - ULHPC
http://hdl.handle.net/10993/40854

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Limited access
Article.pdfAuthor postprint1.16 MBRequest a copy

Additional material(s):

File Commentary Size Access
Limited access
CARLA-NBFL.pdf8.91 MBRequest a copy

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.