A SMOOTHING NEWTON METHOD FOR TENSOR EIGENVALUE COMPLEMENTARITY PROBLEMS

作者：Hu, Wenyu^*; Lu, Laishui; Yin, Cheng; Yu, Gaohang

来源：Pacific Journal of Optimization, 2017, 13(2): 243-253.

• Tensor eigenvalue complementarity problems
• smoothing Newton method
• NCP-function

This paper investigates a smoothing Newton method for tensor eigenvalue complementarity problems (TEiCP) which are closely related to the optimality conditions for polynomial optimization and have wide applications. By introducing a smoothing approximation of NCP-function, the TEiCP problem can be reformulated as a system of smooth equations. Then, a smoothing Newton method is proposed for tensor eigenvalue complementarity problems (TEiCP). Its convergence and convergence rate could be guaranteed by existing results. Numerical experiments are reported to show that the proposed method is efficient and could detect more solutions than some existing optimization-based methods.

• 出版日期2017-4
• 单位江西理工大学; 赣南师范大学