Study of a New Neuron

dc.creatorAdler, S. L.
dc.creatorBhanot, G. V.
dc.creatorWeckel, J. D.
dc.date1994-06-20
dc.date.accessioned2026-07-07T09:05:29Z
dc.date.available2026-07-07T09:05:29Z
dc.descriptionWe study a modular neuron alternative to the McCulloch-Pitts neuron that arises naturally in analog devices in which the neuron inputs are represented as coherent oscillatory wave signals. Although the modular neuron can compute $XOR$ at the one neuron level, it is still characterized by the same Vapnik-Chervonenkis dimension as the standard neuron. We give the formulas needed for constructing networks using the new neuron and training them using back-propagation. A numerical study of the modular neuron on two data sets is presented, which demonstrates that the new neuron performs at least as well as the standard neuron.
dc.description19 pages, 7 figures(not included) available upon request, postscript file
dc.identifierhttps://arxiv.org/abs/adap-org/9406001
dc.identifierhttp://arxiv.org/abs/adap-org/9406001
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/149692
dc.subjectAdaptation and Self-Organizing Systems
dc.subjectQuantitative Biology
dc.titleStudy of a New Neuron
dc.typetext

Files

Collections