Recurrent neural network for approximate nonnegative matrix factorization

详细信息    查看全文

• 作者：Giovanni Costantini^a ; ^{costanti@uniroma2.it" class="auth_mail}Author Vitae ; Renzo Perfetti^b ; ^{renzo.perfetti@unipg.it" class="auth_mail}Author Vitae ; Massimiliano Todisco^a ; ^{massimiliano.todisco@uniroma2.it" class="auth_mail}Author Vitae
• 关键词：Recurrent neural networks ; Lagrangian networks ; Nonnegative matrix factorization ; Features extraction ; Clustering
• 刊名：Neurocomputing
• 出版年：22 August 2014
• 年：2014
• 卷：138
• 期：Complete
• 页码：238-247
• 全文大小：1265 K

文摘

A recurrent neural network solving the approximate nonnegative matrix factorization (NMF) problem is presented in this paper. The proposed network is based on the Lagrangian approach, and exploits a partial dual method in order to limit the number of dual variables. Sparsity constraints on basis or activation matrices are included by adding a weighted sum of constraint functions to the least squares reconstruction error. However, the corresponding Lagrange multipliers are computed by the network dynamics itself, avoiding empirical tuning or a validation process. It is proved that local solutions of the NMF optimization problem correspond to as many stable steady-state points of the network dynamics. The validity of the proposed approach is verified through several simulation examples concerning both synthetic and real-world datasets for feature extraction and clustering applications.