Compressive Spectral Clustering - Inria EPFL Access content directly
Conference Papers Year : 2016

Compressive Spectral Clustering

Abstract

Spectral clustering has become a popular technique due to its high performance in many contexts. It comprises three main steps: create a similarity graph between N objects to cluster, compute the first k eigenvectors of its Laplacian matrix to define a feature vector for each object, and run k-means on these features to separate objects into k classes. Each of these three steps becomes computationally intensive for large N and/or k. We propose to speed up the last two steps based on recent results in the emerging field of graph signal processing: graph filtering of random signals , and random sampling of bandlimited graph signals. We prove that our method, with a gain in computation time that can reach several orders of magnitude, is in fact an approximation of spectral clustering, for which we are able to control the error. We test the performance of our method on artificial and real-world network data.
Fichier principal
Vignette du fichier
ICML_paper.pdf (459.92 Ko) Télécharger le fichier
comptime_vs_N_and_k.pdf (7.19 Ko) Télécharger le fichier
hetero_rec_vs_d.pdf (10.83 Ko) Télécharger le fichier
hetero_rec_vs_gamma.pdf (10.61 Ko) Télécharger le fichier
hetero_rec_vs_n.pdf (10.74 Ko) Télécharger le fichier
hetero_rec_vs_ordPA.pdf (10.28 Ko) Télécharger le fichier
lambda_estimated_N=1000.pdf (8.73 Ko) Télécharger le fichier
rec_vs_N_and_k.pdf (8.22 Ko) Télécharger le fichier
rec_vs_d.pdf (10.81 Ko) Télécharger le fichier
rec_vs_gamma.pdf (10.48 Ko) Télécharger le fichier
rec_vs_n.pdf (10.83 Ko) Télécharger le fichier
rec_vs_ordPA.pdf (10.28 Ko) Télécharger le fichier
table_amazon.pdf (16.99 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)

Dates and versions

hal-01320214 , version 1 (23-05-2016)

Identifiers

Cite

Nicolas Tremblay, Gilles Puy, Rémi Gribonval, Pierre Vandergheynst. Compressive Spectral Clustering. 33rd International Conference on Machine Learning, Jun 2016, New York, United States. ⟨hal-01320214⟩

Relations

627 View
524 Download

Altmetric

Share

Gmail Facebook X LinkedIn More