• TUCS Publication Series
• How to Publish in TUCS Series
• Publication Search
• Publication Graph
• Publication Input
• Publication Input Guide
• JuFo Browser

On Unsupervised Training of Multi-Class Regularized Least-Squares Classifiers

Tapio Pahikkala, Antti Airola, Fabian Gieseke, Oliver Kramer, On Unsupervised Training of Multi-Class Regularized Least-Squares Classifiers. Journal of Computer Science and Technology 29(1), 90–104, 2014.

http://dx.doi.org/10.1007/s11390-014-1414-0

Abstract:

In this work we present the first efficient algorithm for unsupervised training of multi-class regularized least-squares classifiers. The approach is closely related to the unsupervised extension of the support vector machine classifier known as maximum margin clustering, which recently has received considerable attention, though mostly considering the binary classification case. We present a combinatorial search scheme that combines steepest descent strategies with powerful meta-heuristics for avoiding bad local optima. The regularized least-squares based formulation of the problem allows us to use matrix algebraic optimization enabling constant time checks for the intermediate candidate solutions during the search. Our experimental evaluation indicates the potential of the novel method and demonstrates its superior clustering performance over a variety of competing methods on real world datasets. Both time complexity analysis and experimental comparisons show that the method can scale well to practical sized problems.

BibTeX entry:

@ARTICLE{jPaAiGiKr14a,
  title = {On Unsupervised Training of Multi-Class Regularized Least-Squares Classifiers},
  author = {Pahikkala, Tapio and Airola, Antti and Gieseke, Fabian and Kramer, Oliver},
  journal = {Journal of Computer Science and Technology},
  volume = {29},
  number = {1},
  pages = {90–104},
  year = {2014},
  keywords = {Machine learning, kernel methods, clustering, unsupervise classification, ridge regression},
}

Belongs to TUCS Research Unit(s): Algorithmics and Computational Intelligence Group (ACI)

Publication Forum rating of this publication: level 1

Edit publication