Title :
Computing the Modified Hopfield Network Parameters by STFT
Author :
Bauk, Sanja I. ; Perovich, Slavica M.
Author_Institution :
Fac. of Maritime Studies, Montenegro Univ., Kotor
Abstract :
Hopfield has shown that models of physical systems could be used to solve some difficult computational problems. Such systems could be implemented in hardware by combining standard components such as capacitors and resistors. Our idea is to replace capacitor with an inverse polarized diode with large capacity in aim to shift, in a way, Hopfield network parameters analysis from the field of the linear differential equations to the field of the exponential transcendental equations, solvable analytically by the linear approximation method based upon STFT (special trans function theory)
Keywords :
Hopfield neural nets; approximation theory; neural chips; Hopfield network parameter analysis; Hopfield neural network modification; capacitor replacement; exponential transcendental equation; inverse polarized diode; linear approximation method; linear differential equation; special transfunction theory; Capacitors; Computer networks; Differential equations; Diodes; Hopfield neural networks; Linear approximation; Neurons; Polarization; Resistors; Voltage; Hopfield neural network modification; STFT - Special Trans Function Theory based linear approximation method;
Conference_Titel :
Computer as a Tool, 2005. EUROCON 2005.The International Conference on
Conference_Location :
Belgrade
Print_ISBN :
1-4244-0049-X
DOI :
10.1109/EURCON.2005.1629906