Current browse context:
math.OC
Change to browse by:
References & Citations
Mathematics > Optimization and Control
Title: Data Filtering for Cluster Analysis by $\ell_0$-Norm Regularization
(Submitted on 29 Jul 2016 (v1), last revised 22 May 2017 (this version, v3))
Abstract: A data filtering method for cluster analysis is proposed, based on minimizing a least squares function with a weighted $\ell_0$-norm penalty. To overcome the discontinuity of the objective function, smooth non-convex functions are employed to approximate the $\ell_0$-norm. The convergence of the global minimum points of the approximating problems towards global minimum points of the original problem is stated. The proposed method also exploits a suitable technique to choose the penalty parameter. Numerical results on synthetic and real data sets are finally provided, showing how some existing clustering methods can take advantages from the proposed filtering strategy.
Submission history
From: Andrea Cristofari [view email][v1] Fri, 29 Jul 2016 10:21:24 GMT (16kb)
[v2] Fri, 2 Dec 2016 14:23:09 GMT (342kb)
[v3] Mon, 22 May 2017 12:54:58 GMT (344kb)
Link back to: arXiv, form interface, contact.