Logo du site
  • English
  • Français
  • Se connecter
Logo du site
  • English
  • Français
  • Se connecter
  1. Accueil
  2. Université de Neuchâtel
  3. Publications
  4. Minimax-Bayes Reinforcement Learning
 
  • Details
Options
Vignette d'image

Minimax-Bayes Reinforcement Learning

Auteur(s)
Thomas Kleine Buening
Dimitrakakis, Christos 
Institut d'informatique 
Hannes Eriksson
Divya Grover
Emilio Jorge
Maison d'édition
PMLR
Date de parution
2023
In
Proceedings of Machine Learning Research
Vol.
206
De la page
7511
A la page
7527
Mots-clés
  • Machine Learning (cs.LG)
  • Machine Learning (stat.ML)
  • Machine Learning (cs....

  • Machine Learning (sta...

Résumé
While the Bayesian decision-theoretic framework offers an elegant solution to the problem of decision making under uncertainty, one question is how to appropriately select the prior distribution. One idea is to employ a worst-case prior. However, this is not as easy to specify in sequential decision making as in simple statistical estimation problems. This paper studies (sometimes approximate) minimax-Bayes solutions for various reinforcement learning problems to gain insights into the properties of the corresponding priors and policies. We find that while the worst-case prior depends on the setting, the corresponding minimax policies are more robust than those that assume a standard (i.e. uniform) prior.
Notes
International Conference on Artificial Intelligence and Statistics, 25-27 April 2023, Palau de Congressos, Valencia, Spain
Identifiants
https://libra.unine.ch/handle/123456789/30961
_
10.48550/arXiv.2302.10831
Autre version
https://proceedings.mlr.press/v206/buening23a.html
Type de publication
conference paper
Dossier(s) à télécharger
 main article: buening23a.pdf (1.11 MB)
google-scholar
Présentation du portailGuide d'utilisationStratégie Open AccessDirective Open Access La recherche à l'UniNE Open Access ORCIDNouveautés

Service information scientifique & bibliothèques
Rue Emile-Argand 11
2000 Neuchâtel
contact.libra@unine.ch

Propulsé par DSpace, DSpace-CRIS & 4Science | v2022.02.00