Non-negative matrix factorization with L 0 sparseness constraints and its applications to face recognition
Dong Seog Han, Shan Gai · Journal of Intelligent & Fuzzy Systems · 2018
In order to solve the problem of unstable sparseness of non-negative matrix factorization (NMF), the improved NMF algorithms with L 0 sparseness constraints are proposed. With the constraining the L 0 norm of the coefficient matrix, we applied inverse matching principle into non-negative least square (ISNNLS) which enhances the reconstruction ability of the decomposition matrix. In addition, the L 0 sparseness constraints are added to the basis matrix. In the updating process, the proposed algorithm set the smallest value to zero by projecting the basis vectors onto the closest non-negative vector with the expected sparseness. The experimental results have illustrated that the proposed algorithm can achieve higher reconstruction quality and effectiveness compared with the other algorithms.