Loading...

Online Distance Metric Learning

Vazifedan, Afrooz | 2014

743 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46327 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Beigy, Hamid
  7. Abstract:
  8. Distance Metric Learning algorithms have been widely used in Machine Learning methods recently. In these algorithms a distance function between objecs (data points) is learned based on their labels or similarity and dissimilarity constraints. Recent works have shown that a good precision is obtained in classification or clustering methods which use these functions. Since in the current systems many of data points do not exist at the beginning and are added to the training set as the algorithm is run, online methods are needed to update learned metric due to new data.
    In this thesis, we proposed a new online distance metric learning method that has higher performance than existing methods. The proposed method is devised to solve the problem of inseparability of some data sets under a linear transformation. The main idea is to consider different mappings for each class of data points. The proposed method uses adjacent graph in target space in order to close uo similar points and also preserve smoothness. To define target distances, kernel methods has been utilized to increase generalization. In addition, the proposed method is a semi-supervised learning method to also have usage in applications that suffer from shortage of labeled data. Experimental evaluations show higher performance of proposed method in compare with previous works. Theoritical Analysis show that optimization procedure used in each step of the method is a convex one
  9. Keywords:
  10. Online Learning ; Distance Learning ; Semi-Supervised Learning ; Transfer Matrix ; Distance Metric Learning ; Semi-Supervised Classification in Target Space

 Digital Object List

 Bookmark

No TOC