Online allocation with risk information

Shigeaki Harada, Eiji Takimoto, Akira Maruoka

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

1 Citation (Scopus)

Abstract

We consider the problem of dynamically apportioning resources among a set of options in a worst-case online framework. The model we investigate is a generalization of the well studied online learning model. In particular, we allow the learner to see as additional information how high the risk of each option is. This assumption is natural in many applications like horse-race betting, where gamblers know odds for all options before placing bets. We apply the Aggregating Algorithm to this problem and give a tight performance bound. The results support our intuition that we should bet more on low-risk options. Surprisingly, however, the Hedge Algorithm without seeing risk information performs nearly as well as the Aggregating Algorithm. So the risk information does not help much. Moreover, the loss bound does not depend on the values of relatively small risks.

Original languageEnglish
Title of host publicationAlgorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings
Pages343-355
Number of pages13
DOIs
Publication statusPublished - Dec 1 2005
Event16th International Conference on Algorithmic Learning Theory, ALT 2005 - Singapore, Singapore
Duration: Oct 8 2005Oct 11 2005

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3734 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other16th International Conference on Algorithmic Learning Theory, ALT 2005
CountrySingapore
CitySingapore
Period10/8/0510/11/05

Fingerprint

Performance Bounds
Odds
Online Learning
Resources
Model
Generalization
Framework

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Harada, S., Takimoto, E., & Maruoka, A. (2005). Online allocation with risk information. In Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings (pp. 343-355). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3734 LNAI). https://doi.org/10.1007/11564089_27

Online allocation with risk information. / Harada, Shigeaki; Takimoto, Eiji; Maruoka, Akira.

Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings. 2005. p. 343-355 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3734 LNAI).

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

Harada, S, Takimoto, E & Maruoka, A 2005, Online allocation with risk information. in Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3734 LNAI, pp. 343-355, 16th International Conference on Algorithmic Learning Theory, ALT 2005, Singapore, Singapore, 10/8/05. https://doi.org/10.1007/11564089_27
Harada S, Takimoto E, Maruoka A. Online allocation with risk information. In Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings. 2005. p. 343-355. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/11564089_27
Harada, Shigeaki ; Takimoto, Eiji ; Maruoka, Akira. / Online allocation with risk information. Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings. 2005. pp. 343-355 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{a2742a79f11c4aeb806a3ba8e98dbf07,
title = "Online allocation with risk information",
abstract = "We consider the problem of dynamically apportioning resources among a set of options in a worst-case online framework. The model we investigate is a generalization of the well studied online learning model. In particular, we allow the learner to see as additional information how high the risk of each option is. This assumption is natural in many applications like horse-race betting, where gamblers know odds for all options before placing bets. We apply the Aggregating Algorithm to this problem and give a tight performance bound. The results support our intuition that we should bet more on low-risk options. Surprisingly, however, the Hedge Algorithm without seeing risk information performs nearly as well as the Aggregating Algorithm. So the risk information does not help much. Moreover, the loss bound does not depend on the values of relatively small risks.",
author = "Shigeaki Harada and Eiji Takimoto and Akira Maruoka",
year = "2005",
month = "12",
day = "1",
doi = "10.1007/11564089_27",
language = "English",
isbn = "354029242X",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "343--355",
booktitle = "Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings",

}

TY - GEN

T1 - Online allocation with risk information

AU - Harada, Shigeaki

AU - Takimoto, Eiji

AU - Maruoka, Akira

PY - 2005/12/1

Y1 - 2005/12/1

N2 - We consider the problem of dynamically apportioning resources among a set of options in a worst-case online framework. The model we investigate is a generalization of the well studied online learning model. In particular, we allow the learner to see as additional information how high the risk of each option is. This assumption is natural in many applications like horse-race betting, where gamblers know odds for all options before placing bets. We apply the Aggregating Algorithm to this problem and give a tight performance bound. The results support our intuition that we should bet more on low-risk options. Surprisingly, however, the Hedge Algorithm without seeing risk information performs nearly as well as the Aggregating Algorithm. So the risk information does not help much. Moreover, the loss bound does not depend on the values of relatively small risks.

AB - We consider the problem of dynamically apportioning resources among a set of options in a worst-case online framework. The model we investigate is a generalization of the well studied online learning model. In particular, we allow the learner to see as additional information how high the risk of each option is. This assumption is natural in many applications like horse-race betting, where gamblers know odds for all options before placing bets. We apply the Aggregating Algorithm to this problem and give a tight performance bound. The results support our intuition that we should bet more on low-risk options. Surprisingly, however, the Hedge Algorithm without seeing risk information performs nearly as well as the Aggregating Algorithm. So the risk information does not help much. Moreover, the loss bound does not depend on the values of relatively small risks.

UR - http://www.scopus.com/inward/record.url?scp=33646515068&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646515068&partnerID=8YFLogxK

U2 - 10.1007/11564089_27

DO - 10.1007/11564089_27

M3 - Conference contribution

AN - SCOPUS:33646515068

SN - 354029242X

SN - 9783540292425

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 343

EP - 355

BT - Algorithmic Learning Theory - 16th International Conference, ALT 2005, Proceedings

ER -