Minimizing wide range regret with time selection functions

Subhash Khot, Ashok Kumar Ponnuswami

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We consider the problem of minimizing regret with respect to a given set S of pairs of time selection functions and modifications rules. We give an online algorithm that has O(√ T log |S|) regret with respect to S when the algorithm is run for T time steps and there are N actions allowed. This improves the upper bound of O(√ TNlog(|I||F|)) given by Blum and Mansour [BM07a] for the case when S = I × for a set I of time selection functions and a set F of modification rules. We do so by giving a simple reduction that uses an online algorithm for external regret as a black box.

Original languageEnglish (US)
Title of host publication21st Annual Conference on Learning Theory, COLT 2008
Pages81-86
Number of pages6
StatePublished - 2008
Event21st Annual Conference on Learning Theory, COLT 2008 - Helsinki, Finland
Duration: Jul 9 2008Jul 12 2008

Other

Other21st Annual Conference on Learning Theory, COLT 2008
CountryFinland
CityHelsinki
Period7/9/087/12/08

Fingerprint

time

ASJC Scopus subject areas

  • Education

Cite this

Khot, S., & Ponnuswami, A. K. (2008). Minimizing wide range regret with time selection functions. In 21st Annual Conference on Learning Theory, COLT 2008 (pp. 81-86)

Minimizing wide range regret with time selection functions. / Khot, Subhash; Ponnuswami, Ashok Kumar.

21st Annual Conference on Learning Theory, COLT 2008. 2008. p. 81-86.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Khot, S & Ponnuswami, AK 2008, Minimizing wide range regret with time selection functions. in 21st Annual Conference on Learning Theory, COLT 2008. pp. 81-86, 21st Annual Conference on Learning Theory, COLT 2008, Helsinki, Finland, 7/9/08.
Khot S, Ponnuswami AK. Minimizing wide range regret with time selection functions. In 21st Annual Conference on Learning Theory, COLT 2008. 2008. p. 81-86
Khot, Subhash ; Ponnuswami, Ashok Kumar. / Minimizing wide range regret with time selection functions. 21st Annual Conference on Learning Theory, COLT 2008. 2008. pp. 81-86
@inproceedings{14fc196290ef48c49185fb02a46b6cc3,
title = "Minimizing wide range regret with time selection functions",
abstract = "We consider the problem of minimizing regret with respect to a given set S of pairs of time selection functions and modifications rules. We give an online algorithm that has O(√ T log |S|) regret with respect to S when the algorithm is run for T time steps and there are N actions allowed. This improves the upper bound of O(√ TNlog(|I||F|)) given by Blum and Mansour [BM07a] for the case when S = I × for a set I of time selection functions and a set F of modification rules. We do so by giving a simple reduction that uses an online algorithm for external regret as a black box.",
author = "Subhash Khot and Ponnuswami, {Ashok Kumar}",
year = "2008",
language = "English (US)",
pages = "81--86",
booktitle = "21st Annual Conference on Learning Theory, COLT 2008",

}

TY - GEN

T1 - Minimizing wide range regret with time selection functions

AU - Khot, Subhash

AU - Ponnuswami, Ashok Kumar

PY - 2008

Y1 - 2008

N2 - We consider the problem of minimizing regret with respect to a given set S of pairs of time selection functions and modifications rules. We give an online algorithm that has O(√ T log |S|) regret with respect to S when the algorithm is run for T time steps and there are N actions allowed. This improves the upper bound of O(√ TNlog(|I||F|)) given by Blum and Mansour [BM07a] for the case when S = I × for a set I of time selection functions and a set F of modification rules. We do so by giving a simple reduction that uses an online algorithm for external regret as a black box.

AB - We consider the problem of minimizing regret with respect to a given set S of pairs of time selection functions and modifications rules. We give an online algorithm that has O(√ T log |S|) regret with respect to S when the algorithm is run for T time steps and there are N actions allowed. This improves the upper bound of O(√ TNlog(|I||F|)) given by Blum and Mansour [BM07a] for the case when S = I × for a set I of time selection functions and a set F of modification rules. We do so by giving a simple reduction that uses an online algorithm for external regret as a black box.

UR - http://www.scopus.com/inward/record.url?scp=80053150692&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053150692&partnerID=8YFLogxK

M3 - Conference contribution

SP - 81

EP - 86

BT - 21st Annual Conference on Learning Theory, COLT 2008

ER -