New neural network architecture based on quadratic function neurons

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Title of host publicationChina 1991 International Conference on Circuits and Systems
PublisherPubl by IEEE
Pages264-267
ISBN (print)780301502
Publication statusPublished - 1991
Externally publishedYes

Conference

TitleChina 1991 International Conference on Circuits and Systems. Part 1 (of 2)
CityShenzhen, China
Period16 - 17 June 1991

Abstract

In this paper, a class of multilayer perceptron known as rotational quadratic function neural networks (RQFNN) will be introduced. The rotational quadratic function neuron (RQFN), the center stage of this class of networks, is a particular implementation of the quadratic function neuron (QFN). Comparing with the traditional implementation, the RQFN requires much less fan-in's and thus much smaller cross-connection volume. The economy of the fan-in's and the cross connection volumes facilitates the mapping of the model onto silicon.

Citation Format(s)

New neural network architecture based on quadratic function neurons. / Leung, C. S.; Cheung, K. F.; Poon, M. C.
China 1991 International Conference on Circuits and Systems. Publ by IEEE, 1991. p. 264-267.

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review