dc.contributor.author | Calcraft, L. | |
dc.contributor.author | Adams, R.G. | |
dc.contributor.author | Davey, N. | |
dc.date.accessioned | 2011-10-26T14:01:07Z | |
dc.date.available | 2011-10-26T14:01:07Z | |
dc.date.issued | 2006 | |
dc.identifier.citation | Calcraft , L , Adams , R G & Davey , N 2006 , Gaussian and Exponential Architectures in Small-World Associative Memories . in Procs of the European Symposium on Artificial Neural Networks, ESANN'06 . pp. 617-622 , 2006 European Symposium on Artificial Neural Networks, ESANN '06 , Bruges , Belgium , 26/04/06 . | |
dc.identifier.citation | conference | |
dc.identifier.isbn | 2-930307-06-4 | |
dc.identifier.other | PURE: 430974 | |
dc.identifier.other | PURE UUID: b8b4bd8b-52bc-439a-82f3-80041587fbf6 | |
dc.identifier.other | dspace: 2299/2237 | |
dc.identifier.other | Scopus: 78650724683 | |
dc.identifier.uri | http://hdl.handle.net/2299/6803 | |
dc.description.abstract | The performance of sparsely-connected associative memory models built from a set of perceptrons is investigated using different patterns of connectivity. Architectures based on Gaussian and exponential distributions are compared to networks created by progressively rewiring a locally-connected network. It is found that while all three architectures are capable of good pattern-completion performance, the Gaussian and exponential architectures require a significantly lower mean wiring length to achieve the same results. In the case of networks of low connectivity, relatively tight Gaussian and exponential distributions achieve the best overall performance. | en |
dc.language.iso | eng | |
dc.relation.ispartof | Procs of the European Symposium on Artificial Neural Networks, ESANN'06 | |
dc.title | Gaussian and Exponential Architectures in Small-World Associative Memories | en |
dc.contributor.institution | Science & Technology Research Institute | |
dc.contributor.institution | School of Computer Science | |
dc.contributor.institution | Centre for Computer Science and Informatics Research | |
rioxxterms.version | AM | |
rioxxterms.type | Other | |
herts.preservation.rarelyaccessed | true | |