Pure Exploration with Multiple Correct Answers
Rémy Degenne Wouter M. Koolen
Centrum Wiskunde & Informatica Centrum Wiskunde & Informatica
Science Park 123, Amsterdam, NL Science Park 123, Amsterdam, NL
remy.degenne@cwi.nl wmkoolen@cwi.nl
Abstract
We determine the sample complexity of pure exploration bandit problems with
multiple good answers. We derive a lower bound using a new game equili ...


雷达卡



京公网安备 11010802022788号







