Research Article
Dictionary Learning Based on Nonnegative Matrix Factorization Using Parallel Coordinate Descent
Require: Data Matrix , initial matrices , , and ; | (1) while stopping criterion not satisfied do | (2) Computing and ; | (3) for to do | (4) | (5) Normalizing | (6) end for | (7) Computing and ; | (8) for to do | (9) | (10) end for | (11) Using the fixed or adaptively tuning according to the change of the sparsity of ; | (12) end while |
|