1 paper
Dorian Baudry, Kazuya Suzuki, Junya Honda
In this paper we propose a general methodology to derive regret bounds for randomized multi-armed bandit algorithms. It consists in checking a set of sufficient conditions on the s…