Multi-agent systems performance by adaptive/non-adaptive agent selection

Toshiharu Sugawara, Kensuke Fukuda, Toshio Hirotsu, Shin Ya Sato, Satoshi Kurihara

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

3 Citations (Scopus)

Abstract

Our research interest lies in studing how local strategies about partner agent selection using reinforcement learning with variable exploitation-versus- exploration parameters influence the overall efficiency of multi-agent systems (MAS). An agent often has to select appropriate agents to assign tasks that are not locally executable. Unfortunately no agent in an open environment can understand the all states of all agents, so this selection must be done according to local information. In this paper we investigate how the overall performance of MAS is affected by their individual learning parameters for adaptive partner selections for collaboration. We show experimental results using simulation and discuss why the overall performance of MAS varies.

Original languageEnglish
Title of host publicationProceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06
Pages555-559
Number of pages5
DOIs
Publication statusPublished - 2007 Dec 1
Externally publishedYes
Event2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology, IAT'06 - Hong Kong, China
Duration: 2006 Dec 182006 Dec 22

Other

Other2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology, IAT'06
CountryChina
CityHong Kong
Period06/12/1806/12/22

Fingerprint

Multi agent systems
Reinforcement learning

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Software

Cite this

Sugawara, T., Fukuda, K., Hirotsu, T., Sato, S. Y., & Kurihara, S. (2007). Multi-agent systems performance by adaptive/non-adaptive agent selection. In Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06 (pp. 555-559). [4052976] https://doi.org/10.1109/IAT.2006.93

Multi-agent systems performance by adaptive/non-adaptive agent selection. / Sugawara, Toshiharu; Fukuda, Kensuke; Hirotsu, Toshio; Sato, Shin Ya; Kurihara, Satoshi.

Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06. 2007. p. 555-559 4052976.

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

Sugawara, T, Fukuda, K, Hirotsu, T, Sato, SY & Kurihara, S 2007, Multi-agent systems performance by adaptive/non-adaptive agent selection. in Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06., 4052976, pp. 555-559, 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology, IAT'06, Hong Kong, China, 06/12/18. https://doi.org/10.1109/IAT.2006.93
Sugawara T, Fukuda K, Hirotsu T, Sato SY, Kurihara S. Multi-agent systems performance by adaptive/non-adaptive agent selection. In Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06. 2007. p. 555-559. 4052976 https://doi.org/10.1109/IAT.2006.93
Sugawara, Toshiharu ; Fukuda, Kensuke ; Hirotsu, Toshio ; Sato, Shin Ya ; Kurihara, Satoshi. / Multi-agent systems performance by adaptive/non-adaptive agent selection. Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06. 2007. pp. 555-559
@inproceedings{c0b21765721044f1bfc229d5fe3a6533,
title = "Multi-agent systems performance by adaptive/non-adaptive agent selection",
abstract = "Our research interest lies in studing how local strategies about partner agent selection using reinforcement learning with variable exploitation-versus- exploration parameters influence the overall efficiency of multi-agent systems (MAS). An agent often has to select appropriate agents to assign tasks that are not locally executable. Unfortunately no agent in an open environment can understand the all states of all agents, so this selection must be done according to local information. In this paper we investigate how the overall performance of MAS is affected by their individual learning parameters for adaptive partner selections for collaboration. We show experimental results using simulation and discuss why the overall performance of MAS varies.",
author = "Toshiharu Sugawara and Kensuke Fukuda and Toshio Hirotsu and Sato, {Shin Ya} and Satoshi Kurihara",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/IAT.2006.93",
language = "English",
isbn = "9780769527482",
pages = "555--559",
booktitle = "Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06",

}

TY - GEN

T1 - Multi-agent systems performance by adaptive/non-adaptive agent selection

AU - Sugawara, Toshiharu

AU - Fukuda, Kensuke

AU - Hirotsu, Toshio

AU - Sato, Shin Ya

AU - Kurihara, Satoshi

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Our research interest lies in studing how local strategies about partner agent selection using reinforcement learning with variable exploitation-versus- exploration parameters influence the overall efficiency of multi-agent systems (MAS). An agent often has to select appropriate agents to assign tasks that are not locally executable. Unfortunately no agent in an open environment can understand the all states of all agents, so this selection must be done according to local information. In this paper we investigate how the overall performance of MAS is affected by their individual learning parameters for adaptive partner selections for collaboration. We show experimental results using simulation and discuss why the overall performance of MAS varies.

AB - Our research interest lies in studing how local strategies about partner agent selection using reinforcement learning with variable exploitation-versus- exploration parameters influence the overall efficiency of multi-agent systems (MAS). An agent often has to select appropriate agents to assign tasks that are not locally executable. Unfortunately no agent in an open environment can understand the all states of all agents, so this selection must be done according to local information. In this paper we investigate how the overall performance of MAS is affected by their individual learning parameters for adaptive partner selections for collaboration. We show experimental results using simulation and discuss why the overall performance of MAS varies.

UR - http://www.scopus.com/inward/record.url?scp=38949111397&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38949111397&partnerID=8YFLogxK

U2 - 10.1109/IAT.2006.93

DO - 10.1109/IAT.2006.93

M3 - Conference contribution

AN - SCOPUS:38949111397

SN - 9780769527482

SP - 555

EP - 559

BT - Proceedings - 2006 IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2006 Main Conference Proceedings), IAT'06

ER -