A Novel Complex-Valued Fuzzy ARTMAP for Sparse Dictionary Learning

This work extends the simplified fuzzy ARTMAP (SFAM) to a complex-valued (CV-SFAM) version which is able to work with spatio-temporal data produced in receptive fields of visual cortex. The CV-SFAM’s ability for incremental learning distinguishes CV-SFAM from other complex-valued neural networks, wh...

Full description

Saved in:
Bibliographic Details
Published inNeural Information Processing Vol. 8226; pp. 360 - 368
Main Authors Loo, Chu Kiong, Memariani, Ali, Liew, Wei Shiung
Format Book Chapter
LanguageEnglish
Published Germany Springer Berlin / Heidelberg 2013
Springer Berlin Heidelberg
SeriesLecture Notes in Computer Science
Subjects
body expression
complex-valued simplified fuzzy ARTMAP
dictionary learning
genetic optimization
sparse coding
Online AccessGet full text

Cover

More Information
Summary:This work extends the simplified fuzzy ARTMAP (SFAM) to a complex-valued (CV-SFAM) version which is able to work with spatio-temporal data produced in receptive fields of visual cortex. The CV-SFAM’s ability for incremental learning distinguishes CV-SFAM from other complex-valued neural networks, which provides the ability to preserve learned data while learning new samples. We considered different scales and orientations of Gabor wavelets to form a dictionary. This work takes advantage of a locally competitive algorithm (LCA) which calculates more regular sparse coefficients by combining the interactions of artificial neurons. Finally, we provide an experimental real application for biological implementation of sparse dictionary learning to recognize objects in both aligned and non-aligned images.
ISBN:9783642420535
3642420532
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-642-42054-2_45