Loading...
Send to Friend
 Bookmark
 Review List
  1. Simple search
  2. Search result
  3. A new algorithm for learning overcomplete dictionaries

A new algorithm for learning overcomplete dictionaries

Sadeghi, M ; Sharif University of Technology | 2013

764 Viewed
  1. Type of Document: Article
  2. Publisher: European Signal Processing Conference, EUSIPCO , 2013
  3. Abstract:
  4. In this paper, we propose a new algorithm for learning overcomplete dictionaries. The proposed algorithm is actually a new approach for optimizing a recently proposed cost function for dictionary learning. This cost function is regularized with a term that encourages low similarity between different atoms. While the previous approach needs to run an iterative limited-memory BFGS (l-BFGS) algorithm at each iteration of another iterative algorithm, our approach uses a closedform formula. Experimental results on reconstruction of a true underlying dictionary and designing a sparsifying dictionary for a class of autoregressive signals show that our approach results in both better quality and lower computational load
  5. Keywords:
  6. Compressed sensing ; Overcomplete dictionary learning ; Sparse signal approximation ; Cost functions ; Iterative methods ; Signal processing ; Autoregressive signals ; Closed-form formulae ; Computational loads ; Dictionary learning ; Iterative algorithm ; New approaches ; Over-complete dictionaries ; Sparse signals ; Learning algorithms
  7. Source: European Signal Processing Conference, Marrakech ; Sept , 2013 , Page(s): 1 - 4 ; 22195491 (ISSN) ; 9780992862602 (ISBN)
  8. URL: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6811748&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6811748