Fast distributed strategic learning for global optima in queueing access games

Handle URI:
http://hdl.handle.net/10754/575796
Title:
Fast distributed strategic learning for global optima in queueing access games
Authors:
Tembine, Hamidou
Abstract:
In this paper we examine combined fully distributed payoff and strategy learning (CODIPAS) in a queue-aware access game over a graph. The classical strategic learning analysis relies on vanishing or small learning rate and uses stochastic approximation tool to derive steady states and invariant sets of the underlying learning process. Here, the stochastic approximation framework does not apply due to non-vanishing learning rate. We propose a direct proof of convergence of the process. Interestingly, the convergence time to one of the global optima is almost surely finite and we explicitly characterize the convergence time. We show that pursuit-based CODIPAS learning is much faster than the classical learning algorithms in games. We extend the methodology to coalitional learning and proves a very fast formation of coalitions for queue-aware access games where the action space is dynamically changing depending on the location of the user over a graph.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Publisher:
Elsevier BV
Journal:
IFAC Proceedings Volumes
Conference/Event name:
Proceedings of the 19th IFAC World Congress, 2014
Issue Date:
24-Aug-2014
DOI:
10.3182/20140824-6-za-1003.01866
Type:
Conference Paper
ISSN:
14746670
ISBN:
9783902823625
Appears in Collections:
Conference Papers; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorTembine, Hamidouen
dc.date.accessioned2015-08-24T09:26:25Zen
dc.date.available2015-08-24T09:26:25Zen
dc.date.issued2014-08-24en
dc.identifier.isbn9783902823625en
dc.identifier.issn14746670en
dc.identifier.doi10.3182/20140824-6-za-1003.01866en
dc.identifier.urihttp://hdl.handle.net/10754/575796en
dc.description.abstractIn this paper we examine combined fully distributed payoff and strategy learning (CODIPAS) in a queue-aware access game over a graph. The classical strategic learning analysis relies on vanishing or small learning rate and uses stochastic approximation tool to derive steady states and invariant sets of the underlying learning process. Here, the stochastic approximation framework does not apply due to non-vanishing learning rate. We propose a direct proof of convergence of the process. Interestingly, the convergence time to one of the global optima is almost surely finite and we explicitly characterize the convergence time. We show that pursuit-based CODIPAS learning is much faster than the classical learning algorithms in games. We extend the methodology to coalitional learning and proves a very fast formation of coalitions for queue-aware access games where the action space is dynamically changing depending on the location of the user over a graph.en
dc.publisherElsevier BVen
dc.titleFast distributed strategic learning for global optima in queueing access gamesen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalIFAC Proceedings Volumesen
dc.conference.date2014-08-24en
dc.conference.nameProceedings of the 19th IFAC World Congress, 2014en
dc.conference.locationCape Town International Convention Centre, Cape Town, South Africaen
kaust.authorTembine, Hamidouen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.