In this paper, we present an efficient method for nonnegative matrix factorization based on the alternating nonnegative least squares fr a mework. Our approach adopts a monotone projected Barzilai–Borwein (MPBB) method a

子空间 Barzilai-Borwein 梯度算法 对约束问题(Subspace Barzilai-Borwein Gradient Method for Large-Scale Bound Constrained Optimization)