Thompson sampling for one-dimensional ...
Document type :
Communication dans un congrès avec actes
Title :
Thompson sampling for one-dimensional exponential family bandits
Author(s) :
Korda, Nathaniel [Auteur]
Sequential Learning [SEQUEL]
Kaufmann, Emilie [Auteur]
Laboratoire Traitement et Communication de l'Information [LTCI]
Munos, Rémi [Auteur]
Sequential Learning [SEQUEL]
Sequential Learning [SEQUEL]
Kaufmann, Emilie [Auteur]

Laboratoire Traitement et Communication de l'Information [LTCI]
Munos, Rémi [Auteur]
Sequential Learning [SEQUEL]
Conference title :
Advances in Neural Information Processing Systems
Country :
Etats-Unis d'Amérique
Start date of the conference :
2013
Book title :
Advances in Neural Information Processing Systems
Publication date :
2013
HAL domain(s) :
Informatique [cs]/Apprentissage [cs.LG]
English abstract : [en]
Thompson Sampling has been demonstrated in many complex bandit models, however the theoretical guarantees available for the parametric multi-armed bandit are still limited to the Bernoulli case. Here we extend them by ...
Show more >Thompson Sampling has been demonstrated in many complex bandit models, however the theoretical guarantees available for the parametric multi-armed bandit are still limited to the Bernoulli case. Here we extend them by proving asymptotic optimality of the algorithm using the Jeffreys prior for 1-dimensional exponential family bandits. Our proof builds on previous work, but also makes extensive use of closed forms for Kullback-Leibler divergence and Fisher information (and thus Jeffreys prior) available in an exponential family. This allow us to give a finite time exponential concentration inequality for posterior distributions on exponential families that may be of interest in its own right. Moreover our analysis covers some distributions for which no optimistic algorithm has yet been proposed, including heavy-tailed exponential families.Show less >
Show more >Thompson Sampling has been demonstrated in many complex bandit models, however the theoretical guarantees available for the parametric multi-armed bandit are still limited to the Bernoulli case. Here we extend them by proving asymptotic optimality of the algorithm using the Jeffreys prior for 1-dimensional exponential family bandits. Our proof builds on previous work, but also makes extensive use of closed forms for Kullback-Leibler divergence and Fisher information (and thus Jeffreys prior) available in an exponential family. This allow us to give a finite time exponential concentration inequality for posterior distributions on exponential families that may be of interest in its own right. Moreover our analysis covers some distributions for which no optimistic algorithm has yet been proposed, including heavy-tailed exponential families.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-00923683/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-00923683/document
- Open access
- Access the document
- document
- Open access
- Access the document
- nips13-TS.pdf
- Open access
- Access the document
- document
- Open access
- Access the document
- nips13-TS.pdf
- Open access
- Access the document