Loading...

K-LDA: an algorithm for learning jointly overcomplete and discriminative dictionaries

Golmohammady, J ; Sharif University of Technology

551 Viewed
  1. Type of Document: Article
  2. Abstract:
  3. A new algorithm for learning jointly reconstructive and discriminative dictionaries for sparse representation (SR) is presented. While in a usual dictionary learning algorithm like K-SVD only the reconstructive aspect of the sparse representations is considered to learn a dictionary, in our proposed algorithm, which we call K-LDA, the discriminative aspect of the sparse representations is also addressed. In fact, K-LDA is an extension of K-SVD in the case that the class informations (labels) of the training data are also available. K-LDA takes into account these information in order to make the sparse representations more discriminate. It makes a trade-off between the amount of reconstruction error, sparsity, and discrimination of sparse representations. Simulation results on synthetic and hand-written data demonstrate the promising performance of our proposed algorithm
  4. Keywords:
  5. Discriminant analysis ; Economic and social effects ; Signal processing ; Singular value decomposition ; Class information ; Dictionary learning ; Dictionary learning algorithms ; Discriminative dictionaries ; Discriminative learning ; Linear discriminant analysis ; Reconstruction error ; Sparse representation ; Learning algorithms
  6. Source: European Signal Processing Conference ; 10 November 2014 , 2014 , pp. 775-779 ; ISSN: 22195491 ; ISBN: 9780992862619
  7. URL: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6952254&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6952254