DocumentCode
1805981
Title
Hopfield networks
Author
Bemley, Jessye
Author_Institution
St. Francis Xavier Sch., Washington, DC, USA
Volume
6
fYear
1999
fDate
36342
Firstpage
4435
Abstract
A Hopfield network consists of a number of units that are fully connected. Every unit is connected to every other unit. There is a weight associated with each of the inputs that every unit receives from every other unit. Each unit computes the weighted sum of its inputs to generate a net input. The nets were developed by Hopfield (1982). The paper discusses their associative and content-addressable memory, and how they handle changes resulting in differences between the pattern memorised and that presented to them. It considers their energy and their Hamming distance. Their use for optimisation problems is addressed
Keywords
Hamming codes; Hopfield neural nets; learning (artificial intelligence); optimisation; Hamming distance; Hopfield neural networks; associative memory; content-addressable memory; energy; input weighted sum; optimisation; Associative memory; CADCAM; Computer aided manufacturing; Computer networks; History; Hopfield neural networks; Neural networks; Pattern analysis; Pattern recognition; Prototypes;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location
Washington, DC
ISSN
1098-7576
Print_ISBN
0-7803-5529-6
Type
conf
DOI
10.1109/IJCNN.1999.830885
Filename
830885
Link To Document