Size Adaptation of Separable Dictionary Learning with Information-Theoretic Criteria

2019 ◽  
Author(s):  
Andra Baltoiu ◽  
Bogdan Dumitrescu
Keyword(s):  
Dictionary Learning ◽  
Information Theoretic ◽  
Information Theoretic Criteria

scholarly journals Adaptive-Size Dictionary Learning Using Information Theoretic Criteria

Algorithms ◽  
2019 ◽  
Vol 12 (9) ◽  
pp. 178 ◽  
Author(s):  
Bogdan Dumitrescu ◽  
Ciprian Doru Giurcăneanu
Keyword(s):  
Learning Process ◽  
Dictionary Learning ◽  
Numerical Experiments ◽  
Good Accuracy ◽  
Synthetic Data ◽  
Information Theoretic ◽  
Representation Error ◽  
Open Issue ◽  
Information Theoretic Criteria

Finding the size of the dictionary is an open issue in dictionary learning (DL). We propose an algorithm that adapts the size during the learning process by using Information Theoretic Criteria (ITC) specialized to the DL problem. The algorithm is built on top of Approximate K-SVD (AK-SVD) and periodically removes the less used atoms or adds new random atoms, based on ITC evaluations for a small number of candidate sub-dictionaries. Numerical experiments on synthetic data show that our algorithm not only finds the true size with very good accuracy, but is also able to improve the representation error in comparison with AK-SVD knowing the true size.

Sign in / Sign up

Export Citation Format

Share Document