化学计量学中非负单位约束的张量分解算法

In this paper, a tensor CP decomposition model with nonnegative unit constraint is proposed for the measurement of target matter of interest with unknown interference in chemometrics. The proximal Gauss Seidel iterative algorithm is established to decompose the original problem into several subproblems. The generalized polynomial equation and fixed point iteration are applied to solve the subproblems. The convergence theory is established. Finally, the algorithm is numerically simulated and experiments show the effectiveness of the proposed model and algorithm.