DocumentCode :
895069
Title :
Convergence in neural memories
Author :
Dasgupta, Soura ; Ghosh, Anjan ; Cuykendall, Robert
Author_Institution :
Dept. of Electr. Eng., Iowa Univ., Iowa City, IA, USA
Volume :
35
Issue :
5
fYear :
1989
fDate :
9/1/1989 12:00:00 AM
Firstpage :
1069
Lastpage :
1072
Abstract :
One of the simplest optimization problems solved by Ising spin models of neural memory is associative memory retrieval. The authors study deterministic convergence properties of the Hopfield synchronous retrieval algorithm for such models. In this case a memory, stored in the network by an appropriate choice of connections, is retrieved by setting the neural outputs to the binary pattern of the recall key (probe) and allowing the network to converge to a stable state. Precise conditions are developed that ensure that all stored memories are fixed points of the retrieval algorithm. An orthogonality-nearness criterion is then obtained for a memory probe itself to be a stationary point and thus outside the error-correcting capability of the memory. A local stability result quantifies the spatial relationship required for fast convergence
Keywords :
content-addressable storage; convergence; neural nets; Hopfield synchronous retrieval algorithm; Ising spin models; associative memory retrieval; deterministic convergence properties; fast convergence; local stability result; neural memory; neural networks; neurons; optimization problems; orthogonality-nearness criterion; Algorithm design and analysis; Associative memory; Computer networks; Convergence; Integrated circuit interconnections; Marine vehicles; Neural networks; Neurons; Probes; Stability;
fLanguage :
English
Journal_Title :
Information Theory, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9448
Type :
jour
DOI :
10.1109/18.42222
Filename :
42222
Link To Document :
بازگشت