A Game-Theoretic Approach to Competitive Learning in Self-Organizing Maps

Abstract. Self-Organizing Maps (SOM) is a powerful tool for clustering and discovering patterns in data. Competitive learning in the SOM training process focusses on finding a neuron that is most similar to that of an input vector. Since an update of a neuron only benefits part of the feature map, it can be thought of as a local optimization problem. The ability to move away from a local optimization model into a global optimization model requires the use of game theory techniques to analyze overall quality of the SOM. A new algorithm GTSOM is introduced to take into account cluster quality measurements and dynamically modify learning rates to ensure improved quality through successive iterations.

• Type: Conference
• Year: 2005
• Where: ICNC
• Authors: Joseph P. Herbert, Jingtao Yao
• Keywords: ICNC 2005, Local Optimization, Optimization Model, SOM Training Process