Fast Nonnegative Matrix Factorization and Completion Using Nesterov Iterations

详细信息    查看全文

• 关键词：Low ; rank matrix completion ; Nonnegative matrix factorization ; Nesterov iterations ; Gradient descent
• 刊名：Lecture Notes in Computer Science
• 出版年：2017
• 出版时间：2017
• 年：2017
• 卷：10169
• 期：1
• 页码：26-35
• 丛书名：Latent Variable Analysis and Signal Separation
• ISBN：978-3-319-53547-0
• 卷排序：10169

文摘

In this paper, we aim to extend Nonnegative Matrix Factorization with Nesterov iterations (Ne-NMF)—well-suited to large-scale problems—to the situation when some entries are missing in the observed matrix. In particular, we investigate the Weighted and Expectation-Maximization strategies which both provide a way to process missing data. We derive their associated extensions named W-NeNMF and EM-W-NeNMF, respectively. The proposed approaches are then tested on simulated nonnegative low-rank matrix completion problems where the EM-W-NeNMF is shown to outperform state-of-the-art methods and the W-NeNMF technique.