Network of physiological neurons with differentiated excitatory and inhibitory units possessing pattern recognition capacity

Ventouras, E; Kitsonas, M; Hadjiagapis, S; Uzunoglu, N; Papageorgiou, C; Rabavilas, A; Stefanis, C

dc.contributor.author	Ventouras, E	en
dc.contributor.author	Kitsonas, M	en
dc.contributor.author	Hadjiagapis, S	en
dc.contributor.author	Uzunoglu, N	en
dc.contributor.author	Papageorgiou, C	en
dc.contributor.author	Rabavilas, A	en
dc.contributor.author	Stefanis, C	en
dc.date.accessioned	2014-03-01T02:40:59Z
dc.date.available	2014-03-01T02:40:59Z
dc.date.issued	1994	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/30313
dc.subject	Associative Memory	en
dc.subject	Dynamic Behavior	en
dc.subject	Pattern Recognition	en
dc.subject	Neural Network	en
dc.subject.other	Algorithms	en
dc.subject.other	Brain models	en
dc.subject.other	Computer architecture	en
dc.subject.other	Convergence of numerical methods	en
dc.subject.other	Learning systems	en
dc.subject.other	Mathematical models	en
dc.subject.other	Neurophysiology	en
dc.subject.other	Optical correlation	en
dc.subject.other	Parallel processing systems	en
dc.subject.other	Parameter estimation	en
dc.subject.other	Pattern recognition	en
dc.subject.other	Neural populations	en
dc.subject.other	Noise corrupted inputs	en
dc.subject.other	Physiological neurons	en
dc.subject.other	Neural networks	en
dc.title	Network of physiological neurons with differentiated excitatory and inhibitory units possessing pattern recognition capacity	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/NNSP.1994.366047	en
heal.identifier.secondary	http://dx.doi.org/10.1109/NNSP.1994.366047	en
heal.publicationDate	1994	en
heal.abstract	A novel neural network is proposed with different excitatory and inhibitory neural populations. This property conforms with the so-called Dale's hypothesis that applies to neurons of the mammalian brain. The network possesses the ability to store correlated-overlapping patterns and to categorize properly incomplete and noise corrupted inputs. It has the ability to function at realistically low firing rates, such as those found in cortex (15-40 Hz), and to store sparse patterns. Current investigations concern the assessment of the storage capacity of the network and its application in Massively Parallel Computers of the Multiple Instructions-Multiple Data (MIMD) type.	en
heal.publisher	IEEE, Piscataway, NJ, United States	en
heal.journalName	Neural Networks for Signal Processing - Proceedings of the IEEE Workshop	en
dc.identifier.doi	10.1109/NNSP.1994.366047	en
dc.identifier.spage	204	en
dc.identifier.epage	208	en