Title :
A Hopfield neural net and its composition to solve a class of integer programming
Author :
Ae, Tadashi ; Aibara, Reiji ; Agusa, Keiji ; Yamashita, Masafumi
Author_Institution :
Hiroshima Univ., Japan
Abstract :
A special type of parallel processing machine is proposed. The fundamental operation is very fast because it is realized by analog computation. The primitive machine including fundamentally analog computation is a k-state machine, whose operation is realized by the principle of Hopfield neural networks. The total system is constructed by the composition of primitive machines, and can solve a class of polynomially solvable integer programming, with relatively low-cost hardware
Keywords :
analogue simulation; integer programming; neural nets; parallel machines; polynomials; Hopfield neural net; analog computation; cascade connection; integer programming; k-state machine; multiple knapsack problem; parallel connection; parallel processing machine; polynomials; Analog computers; Computer networks; Concurrent computing; Costs; Hopfield neural networks; Linear programming; Neural network hardware; Neural networks; Parallel processing; Polynomials;
Conference_Titel :
Circuits and Systems, 1991., IEEE International Sympoisum on
Print_ISBN :
0-7803-0050-5
DOI :
10.1109/ISCAS.1991.176689
