Loading...

Low-rank kernel learning for semi-supervised clustering

Soleymani Baghshah, M ; Sharif University of Technology | 2010

1072 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/COGINF.2010.5599675
  3. Publisher: 2010
  4. Abstract:
  5. In the last decade, there has been a growing interest in distance function learning for semi-supervised clustering settings. In addition to the earlier methods that learn Mahalanobis metrics (or equivalently, linear transformations), some nonlinear metric learning methods have also been recently introduced. However, these methods either allow limited choice of distance metrics yielding limited flexibility or learn nonparametric kernel matrices and scale very poorly (prohibiting applicability to medium and large data sets). In this paper, we propose a novel method that learns low-rank kernel matrices from pairwise constraints and unlabeled data. We formulate the proposed method as a trace ratio optimization problem and learn appropriate distance metrics through finding optimal low-rank kernel matrices. The proposed optimization problem can be solved much more efficiently than SDP problems introduced to learn nonparametric kernel matrices. Experimental results demonstrate the effectiveness of our method on synthetic and real-world data sets
  6. Keywords:
  7. Low-rank kernel matrix ; Pairwise constraints ; Appropriate distances ; Distance function learning ; Distance metrics ; Kernel learning ; Kernel matrices ; Large datasets ; Mahalanobis ; Metric learning ; Non-parametric ; Novel methods ; Optimization problems ; Pairwise constraints ; Real world data ; Semi-supervised Clustering ; Unlabeled data ; Information science ; Linear transformations ; Optimization ; Matrix algebra
  8. Source: Proceedings of the 9th IEEE International Conference on Cognitive Informatics, ICCI 2010, 7 July 2010 through 9 July 2010, Beijing ; 2010 , Pages 567-572 ; 9781424480401 (ISBN)
  9. URL: http://ieeexplore.ieee.org/document/5599675/?reload=true