BriefGPT.xyz
Aug, 2012
可分非负矩阵分解的快速且鲁棒的递归算法
Fast and Robust Recursive Algorithms for Separable Nonnegative Matrix Factorization
HTML
PDF
Nicolas Gillis, Stephen A. Vavasis
TL;DR
本文研究了非负矩阵分解问题,该问题在可分离性假设下等同于高光谱分离问题。我们提出了一族快速递归算法,并证明了它们对输入数据矩阵的任意小干扰都具有鲁棒性。这一族算法扩展了现有的几种高光谱分离算法,并首次提供了其更好的实际表现的理论证明。
Abstract
In this paper, we study the
nonnegative matrix factorization
problem under the
separability assumption
(that is, there exists a cone spanned by a small subset of the columns of the input nonnegative data matrix c
→