A Bayesian Variational principle for dynamic Self Organizing Maps

TL;DR

This paper introduces a Bayesian variational approach for training self-organizing maps with adaptive neighborhood radii, validated on non-stationary and high-dimensional data, improving their flexibility and performance.

Contribution

It presents a novel Bayesian variational framework for adaptive self-organizing maps, enabling better handling of non-stationary and high-dimensional data.

Findings

Effective in non-stationary environments

Outperforms existing adaptive methods in high-dimensional settings

Demonstrates improved map organization and stability

Abstract

We propose organisation conditions that yield a method for training SOM with adaptative neighborhood radius in a variational Bayesian framework. This method is validated on a non-stationary setting and compared in an high-dimensional setting with an other adaptative method.