Remarks on a recent paper on the "No free lunch" theorems

被引:54
作者
Köppen, M
Wolpert, DH
Macready, WG
机构
[1] Fraunhofer IPK, D-10587 Berlin, Germany
[2] NASA, Ames Res Ctr, Moffett Field, CA 94035 USA
[3] Bios Grp Inc, Santa Fe, NM 87501 USA
关键词
algorithms; genetic algorithms; knowledge representation; learning systems;
D O I
10.1109/4235.930318
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This letter discusses the recent paper "Some technical remarks on the proof of the 'No Free Lunch' theorem," In that paper, some technical issues related to the formal proof of the no free lunch (NFL) theorem for search were given by Wolpert and Macready (1995 and 1997). As a result of a discussion among the authors, this letter explores the issues raised in that paper more thoroughly. This includes the presentation of a simpler version of the NFL proof in accord with a suggestion made explicitly by Koppen (2000) and implicitly by Wolpert and Macready (1997). It also includes the correction of an incorrect claim made by Koppen (2000) of a limitation of the NFL theorem. Finally, some thoughts on future research directions for research into algorithm performance are given.
引用
收藏
页码:295 / 296
页数:2
相关论文
共 6 条
[1]  
Köppen M, 2000, PROCEEDINGS OF THE FIFTH JOINT CONFERENCE ON INFORMATION SCIENCES, VOLS 1 AND 2, P1048
[2]  
WHITLEY LD, 1999, P GEN EV COMP C, V1
[3]  
Wolpert D. H., 1997, IEEE Transactions on Evolutionary Computation, V1, P67, DOI 10.1109/4235.585893
[4]  
Wolpert D.H., 1995, SFITR9502010
[5]   The existence of A priori distinctions between learning algorithms [J].
Wolpert, DH .
NEURAL COMPUTATION, 1996, 8 (07) :1391-1420
[6]   The lack of A priori distinctions between learning algorithms [J].
Wolpert, DH .
NEURAL COMPUTATION, 1996, 8 (07) :1341-1390