Investigation of topographical stability of the concave and convex Self-Organizing Map variant

TL;DR

This paper systematically studies how the stability of the Kohonen Self-Organizing Map and its variants depends on parameters, input distributions, and dimensions through numerical experiments.

Contribution

It provides a comprehensive numerical analysis of the stability conditions for the Kohonen SOM and its concave/convex variants across various settings.

Findings

Stability varies significantly with input distribution and dimensions.

Parameter settings critically influence the stability of the SOM variants.

The study offers guidelines for stable SOM configuration.

Abstract

We investigate, by a systematic numerical study, the parameter dependence of the stability of the Kohonen Self-Organizing Map and the Zheng and Greenleaf concave and convex learning with respect to different input distributions, input and output dimensions.