You are here

FAU Collections » FAU Research Repository » FAU Theses and Dissertations

Self-Organization of Object-Level Visual Representations via Enforcement of Structured Sparsity in Deep Neural Networks

Download pdf | Full Screen View

Date Issued:
2017
Summary:
A hypothesis for the self-organization of receptive fields throughout the hierarchy of biological vision is empirically tested using simulations of deep artificial neural networks. Results from many fields for topographic organization of receptive fields throughout the visual hierarchy remain disconnected. Although extensive simulation research has been done to model topographic organization in early visual areas, little to no research has investigated such organization in higher visual areas. We propose that parsimonious structured sparsity principles, that permit the learning of topographic receptive fields in simulated visual areas, are sufficient for the emergence of a semantic topology in object-level representations of a deep neural network. These findings suggest wide-reaching implications for the functional organization of the biological visual system and we conjecture that such observed results in nature could serve as the foundation for unsupervised learning of taxonomic and semantic relations between entities in the world.
Title: Self-Organization of Object-Level Visual Representations via Enforcement of Structured Sparsity in Deep Neural Networks.
79 views
33 downloads
Name(s): LaCombe, Daniel C. Jr., author
Barenholtz, Elan, Thesis advisor
Florida Atlantic University, Degree grantor
Charles E. Schmidt College of Science
Department of Psychology
Type of Resource: text
Genre: Electronic Thesis Or Dissertation
Date Created: 2017
Date Issued: 2017
Publisher: Florida Atlantic University
Place of Publication: Boca Raton, Fla.
Physical Form: application/pdf
Extent: 89 p.
Language(s): English
Summary: A hypothesis for the self-organization of receptive fields throughout the hierarchy of biological vision is empirically tested using simulations of deep artificial neural networks. Results from many fields for topographic organization of receptive fields throughout the visual hierarchy remain disconnected. Although extensive simulation research has been done to model topographic organization in early visual areas, little to no research has investigated such organization in higher visual areas. We propose that parsimonious structured sparsity principles, that permit the learning of topographic receptive fields in simulated visual areas, are sufficient for the emergence of a semantic topology in object-level representations of a deep neural network. These findings suggest wide-reaching implications for the functional organization of the biological visual system and we conjecture that such observed results in nature could serve as the foundation for unsupervised learning of taxonomic and semantic relations between entities in the world.
Identifier: FA00004955 (IID)
Degree granted: Dissertation (Ph.D.)--Florida Atlantic University, 2017.
Collection: FAU Electronic Theses and Dissertations Collection
Note(s): Includes bibliography.
Subject(s): Dissertations, Academic -- Florida Atlantic University
Held by: Florida Atlantic University Libraries
Sublocation: Digital Library
Links: http://purl.flvc.org/fau/fd/FA00004965
Persistent Link to This Record: http://purl.flvc.org/fau/fd/FA00004955
Use and Reproduction: Copyright © is held by the author, with permission granted to Florida Atlantic University to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Use and Reproduction: http://rightsstatements.org/vocab/InC/1.0/
Host Institution: FAU
Is Part of Series: Florida Atlantic University Digital Library Collections.