Robust non-negative matrix factorization with multiple correntropy-induced hypergraph regularizer
Signal Processing(2020)
摘要
•Our model proposes a novel ensemble manifold regularizer, where multiple correntropy-induced hypergraphs are combined to estimate the underlying data manifold.•By incorporating the proposed regularizer into the original NMF framework, two robust NMF methods are proposed.•The experimental results on noisy image data demonstrate that the proposed method is suporior to the existing methods.
更多查看译文
关键词
Hypergraph,Robust,NMF,Noise
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络