Title :
On the design of dynamic associative neural memories
Author :
Savran, M. Erkan ; Morgül, Ömer
Author_Institution :
Dept. of Electr. & Electron. Eng., Bilkent Univ., Ankara, Turkey
fDate :
5/1/1994 12:00:00 AM
Abstract :
We consider the design problem for a class of discrete-time and continuous-time neural networks. We obtain a characterization of all connection weights that store a given set of vectors into the network, that is, each given vector becomes an equilibrium point of the network. We also give sufficient conditions that guarantee the asymptotic stability of these equilibrium points
Keywords :
content-addressable storage; matrix algebra; neural nets; asymptotic stability; connection matrix; connection weights; continuous time neural networks; design; discrete time neural networks; dynamic associative neural memories; equilibrium point; equilibrium points; sufficient conditions; Associative memory; Asymptotic stability; Computer networks; Concurrent computing; Hopfield neural networks; Neural networks; Neurofeedback; Neurons; Output feedback; Parallel processing;
Journal_Title :
Neural Networks, IEEE Transactions on
