Indexed by:期刊论文

Date of Publication:2015-08-01

Journal:4th International Conference on Numerical Algebra and Scientific Computing (NASC)

Included Journals:SCIE、EI、CPCI-S、Scopus

Volume:93

Issue:1,SI

Page Number:145-158

ISSN No.:0022-0833

Key Words:Cubic spline; Finite minimax problem; Smoothing technique; Stabilized Newton method

Abstract:A spline smoothing stabilized Newton method for finite minimax problems is developed. The spline smoothing technique uses a smooth cubic spline instead of a max function, and, at any fixed point, only a few components of the max function are involved; i.e., it introduces an active set technique, so the proposed method is more efficient for minimizing the maximum function of a large number of complicated functions. Some numerical results comparisons with other methods are also given to show the efficiency of the new method.