Recovering Bandits
Ciara Pike-Burke Steffen Grünewlder
Universitat Pompeu Fabra Lancaster University
Barcelona, Spain Lancaster, UK
c.pikeburke@gmail.com s.grunewalder@lancaster.ac.uk
Abstract
We study the recovering bandits problem, a variant of the stochastic multi-armed
bandit problem where the expected reward of each arm varies according to some
unknown function ...
附件列表