Neural networks as set-valued dynamical systems and the universality of the windowed Fourier transform

DOI10.1007/BF01275642zbMath0833.92003MaRDI QIDQ1898040

Gershon Wolansky

Publication date: 20 September 1995

Published in: Journal of Nonlinear Science (Search for Journal in Brave)

zbMATH Keywords

windowed Fourier transformBorel measurespectral analysisstationary processHebbian learning ruleconvex functionalautocorrelationuniversal attractorGabor filtermultiscale resolutionglobal receptive fieldlayers of neuronsperceptive fieldvisual pathway

Mathematics Subject Classification ID

Neural biology (92C20) Signal detection and filtering (aspects of stochastic processes) (60G35) Other transforms and operators of Fourier type (43A32)

Related Items (1)

A mathematical justification of the grandmother cell hypothesis in neurobiology

Cites Work

• Unnamed Item
• A simplified neuron model as a principal component analyzer
• Self-organization and associative memory.
• Pattern formation in associative neural networks with weak lateral interaction
• Modeling Brain Function
• The Perron–Frobenius Theorem and the Ranking of Football Teams
• Convex Analysis
• A logical calculus of the ideas immanent in nervous activity

This page was built for publication: Neural networks as set-valued dynamical systems and the universality of the windowed Fourier transform