On classes of functions for which No Free Lunch results hold

被引：37

作者：

Igel, C ^{[1
]}

Toussaint, M ^{[1
]}

机构：

[1] Ruhr Univ Bochum, Inst Neuroinformat, D-4630 Bochum, Germany

来源：

INFORMATION PROCESSING LETTERS | 2003年 / 86卷 / 06期

关键词：

combinatorial problems; No Free Lunch; optimization; randomized algorithms;

D O I：

10.1016/S0020-0190(03)00222-9

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A study was performed on no free lunch (NFL) theorems for combinatorial optimization state. Some basic definitions related to the sharpened NFL theorem were presented. Some observations regarding structured search spaces and closure under permutation were discussed.

引用

页码：317 / 321

页数：5

共 6 条

[1] Optimization with randomized search heuristics - the (A)NFL theorem, realistic scenarios, and difficult functions [J].

Droste, S ;

Jansen, T ;

Wegener, I .

THEORETICAL COMPUTER SCIENCE, 2002, 287 (01) :131-144

[2]

Droste S, 1999, GECCO-99: PROCEEDINGS OF THE GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, P833

[3]

FELLER W., 1971, INTRO PROBABILITY TH, V1

[4]

Schumacher C., 2001, Proceedings of the Genetic and Evolutionary Computation Conference GECCO-2001, P565

[5]

Whitley D, 1999, GECCO-99: PROCEEDINGS OF THE GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, P726

[6]

Wolpert D. H., 1997, IEEE Transactions on Evolutionary Computation, V1, P67, DOI 10.1109/4235.585893

← 1 →