Title :
Parallel and synchronous search for combinatory quasi-optimum solutions
Author :
Kakeya, Hideki ; Okabe, Yoichi
Author_Institution :
Res. Center for Adv. Sci. & Technol., Tokyo Univ., Japan
Abstract :
Presents an algorithm which realizes fast search for the solutions of combinatorial optimization problems with digital and parallel processors. By modifying the weight matrix, the proposed model avoids oscillation and realizes energy reduction under the synchronous discrete dynamics. As a result, the proposed model realizes quasi-optimum solutions with much fewer iterations than the existing models
Keywords :
combinatorial mathematics; mathematics computing; optimisation; parallel algorithms; search problems; combinatorial optimization problems; combinatorial quasi-optimum solutions; energy reduction; parallel search; synchronous discrete dynamics; synchronous search; weight matrix; Analytical models; Concurrent computing; Costs; Energy states; Neural networks; Neurons; Paper technology; Parallel architectures; Partitioning algorithms; Simulated annealing;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.616219
