Note on free lunches and cross-validation

Cyril Goutte

doi:10.1162/neco.1997.9.6.1245

Note on free lunches and cross-validation

Cyril Goutte

Research output: Contribution to journal › Journal article › Research

Abstract

The ''no-free-lunch'' theorems (Wolpert & Macready, 1995) have sparked heated debate in the computational learning community. A recent communication (Zhu & Rohwer, 1996) attempts to demonstrate the inefficiency of cross-validation on a simple problem. We elaborate on this result by considering a broader class of cross-validation. When used more strictly, cross-validation can yield the expected results on simple examples.

Original language	English
Journal	Neural Computation
Volume	9
Issue number	6
Pages (from-to)	1246-1249
ISSN	0899-7667
DOIs	https://doi.org/10.1162/neco.1997.9.6.1245
Publication status	Published - 1997

Access to Document

10.1162/neco.1997.9.6.1245

OpenUrl availability

Full text

Cite this

@article{e632a67bbebc43e1822a1681f2771c0f,

title = "Note on free lunches and cross-validation",

abstract = "The ''no-free-lunch'' theorems (Wolpert & Macready, 1995) have sparked heated debate in the computational learning community. A recent communication (Zhu & Rohwer, 1996) attempts to demonstrate the inefficiency of cross-validation on a simple problem. We elaborate on this result by considering a broader class of cross-validation. When used more strictly, cross-validation can yield the expected results on simple examples. ",

author = "Cyril Goutte",

year = "1997",

doi = "10.1162/neco.1997.9.6.1245",

language = "English",

volume = "9",

pages = "1246--1249",

journal = "Neural Computation",

issn = "0899-7667",

publisher = "M I T Press",

number = "6",

}

TY - JOUR

T1 - Note on free lunches and cross-validation

AU - Goutte, Cyril

PY - 1997

Y1 - 1997

N2 - The ''no-free-lunch'' theorems (Wolpert & Macready, 1995) have sparked heated debate in the computational learning community. A recent communication (Zhu & Rohwer, 1996) attempts to demonstrate the inefficiency of cross-validation on a simple problem. We elaborate on this result by considering a broader class of cross-validation. When used more strictly, cross-validation can yield the expected results on simple examples.

AB - The ''no-free-lunch'' theorems (Wolpert & Macready, 1995) have sparked heated debate in the computational learning community. A recent communication (Zhu & Rohwer, 1996) attempts to demonstrate the inefficiency of cross-validation on a simple problem. We elaborate on this result by considering a broader class of cross-validation. When used more strictly, cross-validation can yield the expected results on simple examples.

U2 - 10.1162/neco.1997.9.6.1245

DO - 10.1162/neco.1997.9.6.1245

M3 - Journal article

SN - 0899-7667

VL - 9

SP - 1246

EP - 1249

JO - Neural Computation

JF - Neural Computation

IS - 6

ER -

Note on free lunches and cross-validation

Abstract

Access to Document

OpenUrl availability

Fingerprint

Cite this