Complexity bounds for second-order optimality in unconstrained optimization

被引:62
作者
Cartis, C. [1 ]
Gould, N. I. M. [2 ]
Toint, Ph. L. [3 ]
机构
[1] Univ Edinburgh, Sch Math, Edinburgh EH9 3JZ, Midlothian, Scotland
[2] Rutherford Appleton Lab, Computat Sci & Engn Dept, Chilton OX11 0QX, Oxon, England
[3] FUNDP Univ Namur, Namur Ctr Complex Syst naXys, B-5000 Namur, Belgium
基金
英国工程与自然科学研究理事会;
关键词
Evaluation complexity; Worst-case analysis; Nonconvex optimization; Second-order optimality conditions; CUBIC REGULARIZATION; NEWTONS;
D O I
10.1016/j.jco.2011.06.001
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper examines worst-case evaluation bounds for finding weak minimizers in unconstrained optimization. For the cubic regularization algorithm, Nesterov and Polyak (2006) [15] and Cartis et al. (2010) [3] show that at most 0(epsilon(-3)) iterations may have to be performed for finding an iterate which is within E of satisfying second-order optimality conditions. We first show that this bound can be derived for a version of the algorithm, which only uses one-dimensional global optimization of the cubic model and that it is sharp. We next consider the standard trust-region method and show that a bound of the same type may also be derived for this method, and that it is also sharp in some cases. We conclude by showing that a comparison of the bounds on the worst-case behaviour of the cubic regularization and trust-region algorithms favours the first of these methods. (C) 2011 Elsevier Inc. All rights reserved.
引用
收藏
页码:93 / 108
页数:16
相关论文
共 20 条
[1]  
AGARWAL A., 2009, P 23 ANN C NEUR INF
[2]  
[Anonymous], MPS SIAM SERIES OPTI
[3]  
[Anonymous], 2004, INTRO LECT CONVEX OP
[4]   ON THE COMPLEXITY OF STEEPEST DESCENT, NEWTON'S AND REGULARIZED NEWTON'S METHODS FOR NONCONVEX UNCONSTRAINED OPTIMIZATION PROBLEMS [J].
Cartis, C. ;
Gould, N. I. M. ;
Toint, Ph. L. .
SIAM JOURNAL ON OPTIMIZATION, 2010, 20 (06) :2833-2852
[5]   Trust-region and other regularisations of linear least-squares problems [J].
Cartis, C. ;
Gould, N. I. M. ;
Toint, P. L. .
BIT NUMERICAL MATHEMATICS, 2009, 49 (01) :21-53
[6]  
Cartis C., 2010, NAXYS032010 FUNDP U
[7]  
Cartis C., 2008, 0805 FUNDP U NAM DEP
[8]  
CARTIS C, 2011, 11009 ERGO U ED SCH
[9]   Adaptive cubic regularisation methods for unconstrained optimization. Part II: worst-case function- and derivative-evaluation complexity [J].
Cartis, Coralia ;
Gould, Nicholas I. M. ;
Toint, Philippe L. .
MATHEMATICAL PROGRAMMING, 2011, 130 (02) :295-319
[10]   Adaptive cubic regularisation methods for unconstrained optimization. Part I: motivation, convergence and numerical results [J].
Cartis, Coralia ;
Gould, Nicholas I. M. ;
Toint, Philippe L. .
MATHEMATICAL PROGRAMMING, 2011, 127 (02) :245-295