An Introduction to Gamma-Convergence for Spectral Clustering

Abstract : The problem of clustering is to partition the dataset into groups such that elements belonging to the same group are similar and elements belonging to the different groups are dissimilar. The unsupervised nature of the problem makes it widely applicable and also tough to solve objectively. Clustering in the context of image data is referred to as image segmentation. Distance based methods such as K-means fail to detect the non-globular clusters and hence spectral clustering was proposed to overcome this problem. This method detects the non glob-ular structures by projecting the data set into a subspace, in which the usual clustering methods work well. Gamma convergence is the study of asymptotic behavior of minimizers of a family of minimization problems. Such a limit of minimizers is referred to as the gamma limit. Calculating the gamma limit for various variational problems have been proved useful giving a different algorithm and insights into why existing methods work. In this article, we calculate the gamma limit of the spectral clustering methods, analyze its properties, and compare them with minimum spanning tree based clustering methods and spectral clustering methods.
Type de document :
Communication dans un congrès
Digital Geometry and Computer Imagery, Sep 2017, Vienna, Austria
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-01427957
Contributeur : Aditya Challa <>
Soumis le : mardi 27 juin 2017 - 11:08:30
Dernière modification le : mercredi 5 juillet 2017 - 15:26:45

Fichier

GammaLimitSpectralClustering.p...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01427957, version 2

Collections

Citation

Aditya Challa, Sravan Danda, B Daya Sagar, Laurent Najman. An Introduction to Gamma-Convergence for Spectral Clustering. Digital Geometry and Computer Imagery, Sep 2017, Vienna, Austria. <hal-01427957v2>

Partager

Métriques

Consultations de
la notice

31

Téléchargements du document

52