Self Organizing Map algorithm and distortion measure - Archive ouverte HAL
Article Dans Une Revue Neural Networks Année : 2006

Self Organizing Map algorithm and distortion measure

Résumé

We study the statistical meaning of the minimization of distortion measure and the relation between the equilibrium points of the SOM algorithm and the minima of distortion measure. If we assume that the observations and the map lie in an compact Euclidean space, we prove the strong consistency of the map which almost minimizes the empirical distortion. Moreover, after calculating the derivatives of the theoretical distortion measure, we show that the points minimizing this measure and the equilibria of the Kohonen map do not match in general. We illustrate, with a simple example, how this occurs.
Fichier principal
Vignette du fichier
extended.pdf (207.43 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00257839 , version 1 (20-02-2008)

Identifiants

Citer

Joseph Rynkiewicz. Self Organizing Map algorithm and distortion measure. Neural Networks, 2006, 19 (6-7), pp.830-837. ⟨10.1016/j.neunet.2006.05.016⟩. ⟨hal-00257839⟩
78 Consultations
297 Téléchargements

Altmetric

Partager

More