An Online Learning Approach to a Multi-player N-armed Functional Bandit

S O’Neill; O Bagdasar; Antonio Liotta

doi:10.1007/978-3-030-40616-5_41

Back

An Online Learning Approach to a Multi-player N-armed Functional Bandit

Conference proceeding

Peer reviewed

An Online Learning Approach to a Multi-player N-armed Functional Bandit

S O’Neill, O Bagdasar and Antonio Liotta

Numerical Computations: Theory and Algorithms: Third International Conference, NUMTA 2019, Crotone, Italy, June 15–21, 2019, Revised Selected Papers, Part II, Vol.11974, pp.438-445

Lecture Notes in Computer Science, 11974

Triennial International Conference and Summer School on Numerical Computations: Theory and Algorithms, NUMTA 2019 (Crotone, 15/06/2020 - 21/06/2020)

2020

DOI: https://doi.org/10.1007/978-3-030-40616-5_41

Handle:

https://hdl.handle.net/10863/18725

Abstract

Congestion games

Online learning

Multi-armed bandit

Congestion games possess the property of emitting at least one pure Nash equilibrium and have a rich history of practical use in transport modelling. In this paper we approach the problem of modelling equilibrium within congestion games using a decentralised multi-player probabilistic approach via stochastic bandit feedback. Restricting the strategies available to players under the assumption of bounded rationality, we explore an online multiplayer exponential weights algorithm for unweighted atomic routing games and compare this with a ϵ-greedy algorithm.

Files and links (1)

url

https://link.springer.com/chapter/10.1007/978-3-030-40616-5_41View

Details

Title: An Online Learning Approach to a Multi-player N-armed Functional Bandit
Creators: S O’Neill
O Bagdasar
Antonio Liotta
Publication Details: Numerical Computations: Theory and Algorithms: Third International Conference, NUMTA 2019, Crotone, Italy, June 15–21, 2019, Revised Selected Papers, Part II, Vol.11974, pp.438-445
Editor(s): Sergeyev YD, Kvasov DE
ISBN: 9783030406158
EISBN: 9783030406165
ISSN: 0302-9743
EISSN: 1611-3349
Conference: Triennial International Conference and Summer School on Numerical Computations: Theory and Algorithms, NUMTA 2019 (Crotone, 15/06/2020 - 21/06/2020)
Series / Volume: Lecture Notes in Computer Science
11974
Publisher: Springer
Cham
Number of pages: 8
Identifiers: 978-3-030-40615-8
(UNIBZ)36225100
991006146706401241
Web of Science ID: 000593968900041
Scopus ID: 2-s2.0-85080879142
Academic Unit: Faculty of Computer Science
Language: English
Resource Type: Conference proceeding
Author Names String: O’Neill S, Bagdasar O, Liotta A
Additional Description: Editors/Supervisors: Sergeyev YD, Kvasov DE

Metrics

5 Record Views