求解非负矩阵分解的交替非负最小二乘法的一种修正策略
A MODIFIED STRATEGY IN ALTERNATING NON-NEGATIVE LEAST SQUARES FOR NON-NEGATIVE MATRIX FACTORIZATION

本文研究了关于求解非负矩阵分解的交替非负最小二乘法的全局收敛性.利用一种修正策略保证了极限点的存在性，得到了极限点为非负矩阵分解问题的稳定点.此外，给出了推广的修正策略.数值实验结果表明上述修正策略是有效的.
In this paper, we study the global convergence of non-negative least squares (ANLS) for NMF. By applying a modified strategy to guarantee the existence of the limit point, we obtain the limit point is a stationary point of NMF. In addition, we give generalized modified strategies. Numerical experimental results show the above strategies are effective