Sparse Distributed Memory -- Pattern Data Analysis

Type

conference paper

Language

english

Authors

Grebeníček František, Ing. (FEECS BUT)

URL

Keywords

Sparse Distributed Memory; neural network; pattern recognition

Abstract

This paper discusses, how some statistical properties of pattern data can affect efficiency of Kanerva's Sparse Distributed Memory (SDM). Then, it suggests an method which should improve SDM efficiency. The method looks for optimal SDM parameters according to properties of pattern data. Results of simple experiment are included.

Annotation

This paper discusses, how some statistical properties of pattern data can affect efficiency of Kanerva's Sparse Distributed Memory (SDM). Then, it suggests an method which should improve SDM efficiency. The method looks for optimal SDM parameters according to properties of pattern data. Results of simple experiment are included.

Published

2000

Pages

165-170

Proceedings

MOSIS 2000 Proceedings

Conference

34th Spring International Conference on Modeling and Simulation of Systems MOSIS 2000, Rožnov pod Radhoštěm, CZ

ISBN

80-85988-44-5

Place

Rožnov pod Radhoštěm, CZ

BibTeX

@INPROCEEDINGS{FITPUB5789,
   author = "Franti\v{s}ek Greben\'{i}\v{c}ek",
   title = "Sparse Distributed Memory -- Pattern Data Analysis",
   pages = "165--170",
   booktitle = "MOSIS 2000 Proceedings",
   year = 2000,
   location = "Ro\v{z}nov pod Radho\v{s}t\v{e}m, CZ",
   ISBN = "80-85988-44-5",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/5789"
}