Assessing usage patterns to improve data allocation via auctions

R. Azoulay-Schwart, S. Kraus

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

The data allocation problem in incomplete information environments consisting of self-motivated severs responding to users' queries is considered. Periodically, the servers use auctions for allocation of new data items, and for reallocation of old data items. The utility of a server from storing a data item strongly depends on the usage of the item. However, each server has information only about the past usage of the data stored locally, but does not have information about the usage of data stored elsewhere. In this paper we propose that in order to improve the behaviour of the servers in the auctions, each server learns the expected usage of data items from information about past usage of its own data items. We implemented this type of learning process using neural networks. Simulations showed that our learning methods improve the results of the bidding mechanism, and they are better than the results obtained when learning via k-nearest neighbors algorithms.

Original languageEnglish
Title of host publicationProceedings - 4th International Conference on MultiAgent Systems, ICMAS 2000
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages47-54
Number of pages8
ISBN (Electronic)0769506259, 9780769506258
DOIs
StatePublished - 2000
Event4th International Conference on MultiAgent Systems, ICMAS 2000 - Boston, United States
Duration: 10 Jul 200012 Jul 2000

Publication series

NameProceedings - 4th International Conference on MultiAgent Systems, ICMAS 2000

Conference

Conference4th International Conference on MultiAgent Systems, ICMAS 2000
Country/TerritoryUnited States
CityBoston
Period10/07/0012/07/00

Bibliographical note

Publisher Copyright:
© 2000 IEEE.

Fingerprint

Dive into the research topics of 'Assessing usage patterns to improve data allocation via auctions'. Together they form a unique fingerprint.

Cite this