Multivariable Approximation by Convolutional Kernel Networks

0462912 - ÚI 2017 RIV DE eng C - Conference Paper (international conference)
Kůrková, Věra
Multivariable Approximation by Convolutional Kernel Networks.
Proceedings ITAT 2016: Information Technologies - Applications and Theory. Aachen & Charleston: Technical University & CreateSpace Independent Publishing Platform, 2016 - (Brejová, B.), s. 118-122. CEUR Workshop Proceedings, V-1649. ISBN 978-1-5370-1674-0. ISSN 1613-0073.
[ITAT 2016. Conference on Theory and Practice of Information Technologies /16./. Tatranské Matliare (SK), 15.09.2016-19.09.2016]
R&D Projects: GA ČR GA15-18108S
Institutional support: RVO:67985807
Keywords : kernel networks * approximation of functions * Fourier transform
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
http://ceur-ws.org/Vol-1649/118.pdf

Computational units induced by convolutional kernels together with biologically inspired perceptrons belong to the most widespread types of units used in neurocomputing. Radial convolutional kernels with varying widths form RBF (radial-basis-function) networks and these kernels with fixed widths are used in the SVM (support vector machine) algorithm. We investigate suitability of various convolutional kernel units for function approximation. We show that properties of Fourier transforms of convolutional kernels determine whether sets of input-output functions of networks with kernel units are large enough to be universal approximators. We compare these properties with conditions guaranteeing positive semidefinitness of convolutional kernels.
Permanent Link: http://hdl.handle.net/11104/0262258