Title :
Algorithm for a Tabu — Ant Colony Optimizer
Author :
Haynes, David D. ; Corns, Steven M.
Abstract :
A novel Ant inspired method is introduced in which both positive and negative pheromones are used to guide the ant´s selection process. The negative pheromone serves to influence the decision (much like a tabu search) to discourage the exploration of known bad paths. The positive pheromone serves to attract ants to known good paths (as in any conventional ACO.) Psuedocode for the new algorithm is provided. The dual-pheromone, Tabu-ACO is tested against a classic (positive pheromone only) ACO and the results compared. The Prize Collecting Steiner Tree problem is used to benchmark results.
Keywords :
Benchmark testing; Chemicals; Convergence; History; Standards; Steiner trees; Vegetation; animals; artificial intelligence; biological system modeling; cybernetics; distributed control; genetic algorithms; optimization methods; parallel algorithms; problem solving; stochastic automata; trees (graphs);
Conference_Titel :
Evolutionary Computation (CEC), 2015 IEEE Congress on
Conference_Location :
Sendai, Japan
DOI :
10.1109/CEC.2015.7256935
