FOCS
2007
IEEE
13 years 11 months ago
2007
IEEE
We consider a variant of the classic multi-armed bandit problem (MAB), which we call FEEDBACK MAB, where the reward obtained by playing each of n independent arms varies according...