Title :
A novel model of autoassociative memory and its self-organization
Author :
Matsuoka, Kiyotoshi
Author_Institution :
Div. of Control Eng., Kyushu Inst. of Technol., Japan
Abstract :
A network is presented which embodies the orthogonal type of association. A remarkable point of the network is that the weights of the connections between neurons can be determined directly from the correlation matrix derived from the prototype patterns, requiring no pseudoinverse calculation. As a result, the connection weights can also be obtained by an unsupervised, local learning procedure based on the conventional Hebbian principle
Keywords :
content-addressable storage; learning systems; matrix algebra; Hebbian principle; autoassociative memory; connection weights; correlation matrix; local learning; model; neurons connection; self-organization; Backpropagation; Biological neural networks; Brain; Hopfield neural networks; Indium tin oxide; Learning systems; Neural networks; Organizing; Pattern recognition; Robots;
Journal_Title :
Systems, Man and Cybernetics, IEEE Transactions on
