One Swarm per Queen: A Particle Swarm Learning for Stochastic Games

Alain Tcheukam; Hamidou Tembine

doi:10.1109/SASO.2016.22

One Swarm per Queen: A Particle Swarm Learning for Stochastic Games

Alain Tcheukam, Hamidou Tembine

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

Abstract

This article examines a particle swarm collaborative model-free learning algorithm for approximating equilibria of stochastic games with continuous action spaces. The results support the argument that a simple learning algorithm which consists to explore the continuous action set by means of multi-population of particles can provide a satisfactory solution. A collaborative learning between the particles of the same player takes place during the interactions of the game, in which the players and the particles have no direct knowledge of the payoff model. Each particle is allowed to observe her own payoff and has only one-step memory. The existing results linking the outcomes to stationary satisfactory set do not apply to this situation because of continuous action space and non-convex local response. We provide a different approach to stochastic differential inclusion for arbitrary number of agents.

Original language	English (US)
Title of host publication	Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016
Editors	Giacomo Cabri, Gauthier Picard, Niranjan Suri
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	144-145
Number of pages	2
ISBN (Electronic)	9781509035342
DOIs	https://doi.org/10.1109/SASO.2016.22
State	Published - Dec 5 2016
Event	10th IEEE International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016 - Augsburg, Germany Duration: Sep 12 2016 → Sep 16 2016

Publication series

Name	Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016

Other

Other	10th IEEE International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016
Country/Territory	Germany
City	Augsburg
Period	9/12/16 → 9/16/16

Keywords

Energy markets
Particle Swarm
Stochastic Games

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Hardware and Architecture
Control and Optimization

Access to Document

10.1109/SASO.2016.22

Cite this

Tcheukam, A., & Tembine, H. (2016). One Swarm per Queen: A Particle Swarm Learning for Stochastic Games. In G. Cabri, G. Picard, & N. Suri (Eds.), Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016 (pp. 144-145). Article 7774397 (Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SASO.2016.22

One Swarm per Queen: A Particle Swarm Learning for Stochastic Games. / Tcheukam, Alain; Tembine, Hamidou.
Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016. ed. / Giacomo Cabri; Gauthier Picard; Niranjan Suri. Institute of Electrical and Electronics Engineers Inc., 2016. p. 144-145 7774397 (Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016).

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

Tcheukam, A & Tembine, H 2016, One Swarm per Queen: A Particle Swarm Learning for Stochastic Games. in G Cabri, G Picard & N Suri (eds), Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016., 7774397, Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016, Institute of Electrical and Electronics Engineers Inc., pp. 144-145, 10th IEEE International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016, Augsburg, Germany, 9/12/16. https://doi.org/10.1109/SASO.2016.22

Tcheukam A, Tembine H. One Swarm per Queen: A Particle Swarm Learning for Stochastic Games. In Cabri G, Picard G, Suri N, editors, Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 144-145. 7774397. (Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016). doi: 10.1109/SASO.2016.22

Tcheukam, Alain ; Tembine, Hamidou. / One Swarm per Queen : A Particle Swarm Learning for Stochastic Games. Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016. editor / Giacomo Cabri ; Gauthier Picard ; Niranjan Suri. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 144-145 (Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016).

@inproceedings{68b8f285e24f4b958ade18d3d796c5d9,

title = "One Swarm per Queen: A Particle Swarm Learning for Stochastic Games",

abstract = "This article examines a particle swarm collaborative model-free learning algorithm for approximating equilibria of stochastic games with continuous action spaces. The results support the argument that a simple learning algorithm which consists to explore the continuous action set by means of multi-population of particles can provide a satisfactory solution. A collaborative learning between the particles of the same player takes place during the interactions of the game, in which the players and the particles have no direct knowledge of the payoff model. Each particle is allowed to observe her own payoff and has only one-step memory. The existing results linking the outcomes to stationary satisfactory set do not apply to this situation because of continuous action space and non-convex local response. We provide a different approach to stochastic differential inclusion for arbitrary number of agents.",

keywords = "Energy markets, Particle Swarm, Stochastic Games",

author = "Alain Tcheukam and Hamidou Tembine",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 10th IEEE International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016 ; Conference date: 12-09-2016 Through 16-09-2016",

year = "2016",

month = dec,

day = "5",

doi = "10.1109/SASO.2016.22",

language = "English (US)",

series = "Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "144--145",

editor = "Giacomo Cabri and Gauthier Picard and Niranjan Suri",

booktitle = "Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016",

}

TY - GEN

T1 - One Swarm per Queen

T2 - 10th IEEE International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016

AU - Tcheukam, Alain

AU - Tembine, Hamidou

PY - 2016/12/5

Y1 - 2016/12/5

N2 - This article examines a particle swarm collaborative model-free learning algorithm for approximating equilibria of stochastic games with continuous action spaces. The results support the argument that a simple learning algorithm which consists to explore the continuous action set by means of multi-population of particles can provide a satisfactory solution. A collaborative learning between the particles of the same player takes place during the interactions of the game, in which the players and the particles have no direct knowledge of the payoff model. Each particle is allowed to observe her own payoff and has only one-step memory. The existing results linking the outcomes to stationary satisfactory set do not apply to this situation because of continuous action space and non-convex local response. We provide a different approach to stochastic differential inclusion for arbitrary number of agents.

AB - This article examines a particle swarm collaborative model-free learning algorithm for approximating equilibria of stochastic games with continuous action spaces. The results support the argument that a simple learning algorithm which consists to explore the continuous action set by means of multi-population of particles can provide a satisfactory solution. A collaborative learning between the particles of the same player takes place during the interactions of the game, in which the players and the particles have no direct knowledge of the payoff model. Each particle is allowed to observe her own payoff and has only one-step memory. The existing results linking the outcomes to stationary satisfactory set do not apply to this situation because of continuous action space and non-convex local response. We provide a different approach to stochastic differential inclusion for arbitrary number of agents.

KW - Energy markets

KW - Particle Swarm

KW - Stochastic Games

UR - http://www.scopus.com/inward/record.url?scp=85010409252&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85010409252&partnerID=8YFLogxK

U2 - 10.1109/SASO.2016.22

DO - 10.1109/SASO.2016.22

M3 - Conference contribution

AN - SCOPUS:85010409252

T3 - Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016

SP - 144

EP - 145

BT - Proceedings - IEEE 10th International Conference on Self-Adaptive and Self-Organizing Systems, SASO 2016

A2 - Cabri, Giacomo

A2 - Picard, Gauthier

A2 - Suri, Niranjan

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 12 September 2016 through 16 September 2016

ER -

One Swarm per Queen: A Particle Swarm Learning for Stochastic Games

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this