Title :
Experience with the reactive tabu search as implemented in the TOTEM chip
Author :
Lindsey, Clark S. ; Lindblad, Thomas
Author_Institution :
Dept. of Phys., R. Inst. of Technol., Stockholm, Sweden
Abstract :
The reactive tabu search (RTS) uses combinatorial optimization and is an algorithm in which no derivative calculations are needed. Like the tabu search (TS), it uses the search experience to guide the search process in order to find high quality solutions to combinatorial optimization problems. It also works with low precision weight, properties which make the RTS suitable for implementation in hardware. In this paper we present some results of the use of the RTS algorithm with the TOTEM chip for applications in high energy physics (Higgs search) and with some benchmarks from the ELENA Esprit project. Comparisons with results from three other circuit are presented
Keywords :
feedforward neural nets; neural chips; optimisation; physics; physics computing; search problems; ELENA Esprit project; Higgs search; TOTEM chip; combinatorial optimization; feedforward neural network; high energy physics; neural chip; reactive tabu search; Circuits; Filtering; Neural network hardware; Neural networks; Physics; Prototypes; Testing;
Conference_Titel :
Neural Networks, 1996., IEEE International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-3210-5
DOI :
10.1109/ICNN.1996.549033