General oracle inequalities for gibbs posterior with application to ranking

Cheng Li, Wenxin Jiang, Martin A Tanner

Research output: Contribution to journalConference article

4 Citations (Scopus)

Abstract

In this paper, we summarize some recent results in Li et al. (2012), which can be used to extend an important PAC-Bayesian approach, namely the Gibbs posterior, to study the nonadditive ranking risk. The methodology is based on assumption-free risk bounds and nonasymptotic oracle inequalities, which leads to nearly optimal convergence rates and optimal model selection to balance the approximation errors and the stochastic errors.

Original languageEnglish (US)
Pages (from-to)512-521
Number of pages10
JournalJournal of Machine Learning Research
Volume30
StatePublished - Jan 1 2013
Event26th Conference on Learning Theory, COLT 2013 - Princeton, NJ, United States
Duration: Jun 12 2013Jun 14 2013

Fingerprint

Oracle Inequalities
Ranking
Optimal Convergence Rate
Approximation Error
Bayesian Approach
Model Selection
Methodology

Keywords

  • Gibbs posterior
  • Model selection
  • Oracle inequalities
  • Ranking
  • Risk minimization

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Statistics and Probability
  • Artificial Intelligence

Cite this

@article{0bb3ded136f74459b0133d9be67a4e9f,
title = "General oracle inequalities for gibbs posterior with application to ranking",
abstract = "In this paper, we summarize some recent results in Li et al. (2012), which can be used to extend an important PAC-Bayesian approach, namely the Gibbs posterior, to study the nonadditive ranking risk. The methodology is based on assumption-free risk bounds and nonasymptotic oracle inequalities, which leads to nearly optimal convergence rates and optimal model selection to balance the approximation errors and the stochastic errors.",
keywords = "Gibbs posterior, Model selection, Oracle inequalities, Ranking, Risk minimization",
author = "Cheng Li and Wenxin Jiang and Tanner, {Martin A}",
year = "2013",
month = "1",
day = "1",
language = "English (US)",
volume = "30",
pages = "512--521",
journal = "Journal of Machine Learning Research",
issn = "1532-4435",
publisher = "Microtome Publishing",

}

General oracle inequalities for gibbs posterior with application to ranking. / Li, Cheng; Jiang, Wenxin; Tanner, Martin A.

In: Journal of Machine Learning Research, Vol. 30, 01.01.2013, p. 512-521.

Research output: Contribution to journalConference article

TY - JOUR

T1 - General oracle inequalities for gibbs posterior with application to ranking

AU - Li, Cheng

AU - Jiang, Wenxin

AU - Tanner, Martin A

PY - 2013/1/1

Y1 - 2013/1/1

N2 - In this paper, we summarize some recent results in Li et al. (2012), which can be used to extend an important PAC-Bayesian approach, namely the Gibbs posterior, to study the nonadditive ranking risk. The methodology is based on assumption-free risk bounds and nonasymptotic oracle inequalities, which leads to nearly optimal convergence rates and optimal model selection to balance the approximation errors and the stochastic errors.

AB - In this paper, we summarize some recent results in Li et al. (2012), which can be used to extend an important PAC-Bayesian approach, namely the Gibbs posterior, to study the nonadditive ranking risk. The methodology is based on assumption-free risk bounds and nonasymptotic oracle inequalities, which leads to nearly optimal convergence rates and optimal model selection to balance the approximation errors and the stochastic errors.

KW - Gibbs posterior

KW - Model selection

KW - Oracle inequalities

KW - Ranking

KW - Risk minimization

UR - http://www.scopus.com/inward/record.url?scp=84898033757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898033757&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:84898033757

VL - 30

SP - 512

EP - 521

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

SN - 1532-4435

ER -