Higher Order Pheromone Models in Ant Colony Optimisation
Ant colony optimisation is a constructive metaheuristic that successively builds solutions from problem-specific components. A parameterised model known as pheromone—an analogue of the trail pheromones used by real ants—is used to learn which components should be combined to produce good solutions. In the majority of the algorithm’s applications a single parameter from the model is used to influence the selection of a single component to add to a solution. Such a model can be described as first order. Higher order models describe relationships between several components in a solution, and may arise either by contriving a model that describes subsets of components from a first order model or because the characteristics of solutions modelled naturally relate subsets of components. This paper introduces a simple framework to describe the application of higher order models as a tool to understanding common features of existing applications. The framework also serves as an introduction to those new to the use of such models. The utility of higher order models is discussed with reference to empirical results in the literature.
KeywordsOrder Model Travel Salesman Problem Partial Solution Constraint Satisfaction Problem Solution Component
Unable to display preview. Download preview PDF.
- 3.Blum, C.: Theoretical and practical aspects of ant colony optimization. Dissertations in Artificial Intelligence, vol. 282. IOS Press, Nieuwe Hemweg (2004)Google Scholar
- 4.Blum, C., Sampels, M.: Ant colony optimization for FOP shop scheduling: A case study on different pheromone representations. In: 2002 Congress on Evolutionary Computation, pp. 1558–1563 (2002)Google Scholar
- 5.Montgomery, E.J.: Solution Biases and Pheromone Representation Selection in Ant Colony Optimisation. Ph.D thesis, Bond University (2005)Google Scholar
- 7.Roli, A., Blum, C., Dorigo, M.: ACO for maximal constraint satisfaction problems. In: 4th Metaheuristics International Conference, Porto, Portugal, pp. 187–192 (2001)Google Scholar
- 8.Solnon, C.: Ants can solve constraint satisfaction problems. IEEE Trans. Evol. Comput. 6 (2002)Google Scholar
- 9.Fenet, S., Solnon, C.: Searching for maximum cliques with ant colony optimization. In: Raidl, G.R., Cagnoni, S., Cardalda, J.J.R., Corne, D.W., Gottlieb, J., Guillot, A., Hart, E., Johnson, C.G., Marchiori, E., Meyer, J.-A., Middendorf, M. (eds.) EvoIASP 2003, EvoWorkshops 2003, EvoSTIM 2003, EvoROB/EvoRobot 2003, EvoCOP 2003, EvoBIO 2003, and EvoMUSART 2003. LNCS, vol. 2611, pp. 236–245. Springer, Heidelberg (2003)CrossRefGoogle Scholar
- 10.Blum, C.: Ant colony optimization for the edge-weighted k-cardinality tree problem. In: Langdon, W. (ed.) Genetic and Evolutionary Computation Conference, pp. 27–34. Morgan Kaufmann Publishers, San Francisco (2002)Google Scholar
- 12.Schoofs, L., Naudts, B.: Solving CSPs with ant colonies. In: Abstract Proceedings of ANTS 2000, Brussels, Belgium (2000)Google Scholar