Towards robust model selection using estimation and approximation error bounds

Joel Ratsaby; Ronny Meir; Vitaly Maiorov

doi:10.1145/238061.238069

Towards robust model selection using estimation and approximation error bounds

Joel Ratsaby, Ronny Meir, Vitaly Maiorov

Research output: Contribution to conference › Paper › peer-review

6 Scopus citations

Abstract

One of the main problems in machine learning and statistical inference is selecting an appropriate model by which a set of data can be explained. A novel model selection criterion based on the uniform convergence of empirical processes combined with the results concerning the approximation ability of non-linear manifolds of functions is introduced. A coherent and robust framework for model selection was elucidated and a lower bound on the approximation error was established, giving a well specified sense for most functions of interest.

Original language	English
Pages	57-67
Number of pages	11
DOIs	https://doi.org/10.1145/238061.238069
State	Published - 1 Jan 1996
Externally published	Yes
Event	Proceedings of the 1996 9th Annual Conference on Computational Learning Theory - Desenzano del Garda, Italy Duration: 28 Jun 1996 → 1 Jul 1996

Conference

Conference	Proceedings of the 1996 9th Annual Conference on Computational Learning Theory
City	Desenzano del Garda, Italy
Period	28/06/96 → 1/07/96

ASJC Scopus subject areas

Computational Mathematics

Access to Document

10.1145/238061.238069

Cite this

@conference{150263fbf3984023b30e03c614f2c665,

title = "Towards robust model selection using estimation and approximation error bounds",

abstract = "One of the main problems in machine learning and statistical inference is selecting an appropriate model by which a set of data can be explained. A novel model selection criterion based on the uniform convergence of empirical processes combined with the results concerning the approximation ability of non-linear manifolds of functions is introduced. A coherent and robust framework for model selection was elucidated and a lower bound on the approximation error was established, giving a well specified sense for most functions of interest.",

author = "Joel Ratsaby and Ronny Meir and Vitaly Maiorov",

year = "1996",

month = jan,

day = "1",

doi = "10.1145/238061.238069",

language = "English",

pages = "57--67",

note = "Proceedings of the 1996 9th Annual Conference on Computational Learning Theory ; Conference date: 28-06-1996 Through 01-07-1996",

}

TY - CONF

T1 - Towards robust model selection using estimation and approximation error bounds

AU - Ratsaby, Joel

AU - Meir, Ronny

AU - Maiorov, Vitaly

PY - 1996/1/1

Y1 - 1996/1/1

N2 - One of the main problems in machine learning and statistical inference is selecting an appropriate model by which a set of data can be explained. A novel model selection criterion based on the uniform convergence of empirical processes combined with the results concerning the approximation ability of non-linear manifolds of functions is introduced. A coherent and robust framework for model selection was elucidated and a lower bound on the approximation error was established, giving a well specified sense for most functions of interest.

AB - One of the main problems in machine learning and statistical inference is selecting an appropriate model by which a set of data can be explained. A novel model selection criterion based on the uniform convergence of empirical processes combined with the results concerning the approximation ability of non-linear manifolds of functions is introduced. A coherent and robust framework for model selection was elucidated and a lower bound on the approximation error was established, giving a well specified sense for most functions of interest.

UR - http://www.scopus.com/inward/record.url?scp=0030383583&partnerID=8YFLogxK

U2 - 10.1145/238061.238069

DO - 10.1145/238061.238069

M3 - Paper

AN - SCOPUS:0030383583

SP - 57

EP - 67

T2 - Proceedings of the 1996 9th Annual Conference on Computational Learning Theory

Y2 - 28 June 1996 through 1 July 1996

ER -

Towards robust model selection using estimation and approximation error bounds

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this