SOMulANT: Organizing Information Using Multiple Agents
This paper introduces an algorithm, inspired by ants, that allows information to be sorted onto a map. The agents that carry out the mapping have vision and an ability to measure the local disorder at points on the map. They use these to decide when to lift information and where and when to place it down. Results of sorting a well-known data set, the iris data set, are presented and the importance of various parameters discussed.
KeywordsCollective intelligence self organizing map information mapping multiple agents
Unable to display preview. Download preview PDF.
- 1.Kohonen T. ‘Self-Organisiation and Associative Memory’ Springer-Verlag New York.Google Scholar
- 2.Deneubourg, J-L, Goss S, Franks N, Sendova-Franks A, Detrain C and Chretien L. ‘The Dynamics of Collective Sorting: Robot-Like Ant and Ant-Like Robot.’ In Proceedings First European Conference on Simulation of Adaptive Behaviour: from Animals to Animats, edited FJ Varela and P Bourgine, pages 123–133. Cambridge MA MIT press 1992.Google Scholar
- 4.Lumer E and Faieta B, ‘Diversity and Adaption in Populations of Clustering Ants’. In Proceedings Third International Conference on Simulation of Adaptive Behaviour: from Animals to Animats, pages 499–508, Cambridge MA MIT Press 1994.Google Scholar
- 5.Batchelor B.G.. ‘Practical Approach to Pattern Recognition’. Plenum Press, New York 1974.Google Scholar