An adaptive-learning framework for semi-cooperative multi-agent coordination

Abdeslem Boukhtouta; Jean Berger; Warren Buckler Powell; Abraham George

doi:10.1109/ADPRL.2011.5967386

An adaptive-learning framework for semi-cooperative multi-agent coordination

Abdeslem Boukhtouta, Jean Berger, Warren Buckler Powell, Abraham George

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Scopus citations

Abstract

Complex problems involving multiple agents exhibit varying degrees of cooperation. The levels of cooperation might reflect both differences in information as well as differences in goals. In this research, we develop a general mathematical model for distributed, semi-cooperative planning and suggest a solution strategy which involves decomposing the system into subproblems, each of which is specified at a certain period in time and controlled by an agent. The agents communicate marginal values of resources to each other, possibly with distortion. We design experiments to demonstrate the benefits of communication between the agents and show that, with communication, the solution quality approaches that of the ideal situation where the entire problem is controlled by a single agent.

Original language	English (US)
Title of host publication	IEEE SSCI 2011
Subtitle of host publication	Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning
Pages	324-331
Number of pages	8
DOIs	https://doi.org/10.1109/ADPRL.2011.5967386
State	Published - 2011
Event	Symposium Series on Computational Intelligence, IEEE SSCI2011 - 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2011 - Paris, France Duration: Apr 11 2011 → Apr 15 2011

Publication series

Name	IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning

Other

Other	Symposium Series on Computational Intelligence, IEEE SSCI2011 - 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2011
Country/Territory	France
City	Paris
Period	4/11/11 → 4/15/11

All Science Journal Classification (ASJC) codes

Computer Science Applications
Software

Keywords

Multi-agent
approximate dynamic programming
cooperative
learning

Access to Document

10.1109/ADPRL.2011.5967386

Cite this

Boukhtouta, A., Berger, J., Powell, W. B., & George, A. (2011). An adaptive-learning framework for semi-cooperative multi-agent coordination. In IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning (pp. 324-331). Article 5967386 (IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning). https://doi.org/10.1109/ADPRL.2011.5967386

Boukhtouta, Abdeslem ; Berger, Jean ; Powell, Warren Buckler et al. / An adaptive-learning framework for semi-cooperative multi-agent coordination. IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning. 2011. pp. 324-331 (IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning).

@inproceedings{d48723b23a484ea9be8c1def1cc45591,

title = "An adaptive-learning framework for semi-cooperative multi-agent coordination",

abstract = "Complex problems involving multiple agents exhibit varying degrees of cooperation. The levels of cooperation might reflect both differences in information as well as differences in goals. In this research, we develop a general mathematical model for distributed, semi-cooperative planning and suggest a solution strategy which involves decomposing the system into subproblems, each of which is specified at a certain period in time and controlled by an agent. The agents communicate marginal values of resources to each other, possibly with distortion. We design experiments to demonstrate the benefits of communication between the agents and show that, with communication, the solution quality approaches that of the ideal situation where the entire problem is controlled by a single agent.",

keywords = "Multi-agent, approximate dynamic programming, cooperative, learning",

author = "Abdeslem Boukhtouta and Jean Berger and Powell, {Warren Buckler} and Abraham George",

year = "2011",

doi = "10.1109/ADPRL.2011.5967386",

language = "English (US)",

isbn = "9781424498888",

series = "IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning",

pages = "324--331",

booktitle = "IEEE SSCI 2011",

note = "Symposium Series on Computational Intelligence, IEEE SSCI2011 - 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2011 ; Conference date: 11-04-2011 Through 15-04-2011",

}

Boukhtouta, A, Berger, J, Powell, WB & George, A 2011, An adaptive-learning framework for semi-cooperative multi-agent coordination. in IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning., 5967386, IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, pp. 324-331, Symposium Series on Computational Intelligence, IEEE SSCI2011 - 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2011, Paris, France, 4/11/11. https://doi.org/10.1109/ADPRL.2011.5967386

An adaptive-learning framework for semi-cooperative multi-agent coordination. / Boukhtouta, Abdeslem; Berger, Jean; Powell, Warren Buckler et al.
IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning. 2011. p. 324-331 5967386 (IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - An adaptive-learning framework for semi-cooperative multi-agent coordination

AU - Boukhtouta, Abdeslem

AU - Berger, Jean

AU - Powell, Warren Buckler

AU - George, Abraham

PY - 2011

Y1 - 2011

N2 - Complex problems involving multiple agents exhibit varying degrees of cooperation. The levels of cooperation might reflect both differences in information as well as differences in goals. In this research, we develop a general mathematical model for distributed, semi-cooperative planning and suggest a solution strategy which involves decomposing the system into subproblems, each of which is specified at a certain period in time and controlled by an agent. The agents communicate marginal values of resources to each other, possibly with distortion. We design experiments to demonstrate the benefits of communication between the agents and show that, with communication, the solution quality approaches that of the ideal situation where the entire problem is controlled by a single agent.

AB - Complex problems involving multiple agents exhibit varying degrees of cooperation. The levels of cooperation might reflect both differences in information as well as differences in goals. In this research, we develop a general mathematical model for distributed, semi-cooperative planning and suggest a solution strategy which involves decomposing the system into subproblems, each of which is specified at a certain period in time and controlled by an agent. The agents communicate marginal values of resources to each other, possibly with distortion. We design experiments to demonstrate the benefits of communication between the agents and show that, with communication, the solution quality approaches that of the ideal situation where the entire problem is controlled by a single agent.

KW - Multi-agent

KW - approximate dynamic programming

KW - cooperative

KW - learning

UR - http://www.scopus.com/inward/record.url?scp=80052228198&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052228198&partnerID=8YFLogxK

U2 - 10.1109/ADPRL.2011.5967386

DO - 10.1109/ADPRL.2011.5967386

M3 - Conference contribution

AN - SCOPUS:80052228198

SN - 9781424498888

T3 - IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning

SP - 324

EP - 331

BT - IEEE SSCI 2011

T2 - Symposium Series on Computational Intelligence, IEEE SSCI2011 - 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2011

Y2 - 11 April 2011 through 15 April 2011

ER -

Boukhtouta A, Berger J, Powell WB, George A. An adaptive-learning framework for semi-cooperative multi-agent coordination. In IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning. 2011. p. 324-331. 5967386. (IEEE SSCI 2011: Symposium Series on Computational Intelligence - ADPRL 2011: 2011 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning). doi: 10.1109/ADPRL.2011.5967386

An adaptive-learning framework for semi-cooperative multi-agent coordination

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this