Parallel Manner and Twist Measurement for Self-Organizing Maps

This paper investigates a self-organizing map algorithm which is simultaneously given multiple inputs, and introduces a criterion of ordering process as the twist measurement for reference vectors in multidimensional array. The self-organizing algorithm and the criterion are termed the parallel self-organizing algorithm and the index of twist, respectively. The algorithm updates reference vectors corresponding to respective inputs at the same time, when multiple inputs are prepared at each step. The index of twist is the criterion to evaluate the multidimensional ordering of the topological array for reference vectors. When the parallel degree changes for the present algorithm, the topology preserving map after learning is evaluated by utilizing the index of twist. By discussing the formation rate and the average distortion of topology preserving map, the effectiveness of the present approach is examined. Key-Words: Neural networks, Self-organizing map, Parallel manner, Twist measurement Learning, Numerical experiment