Improved Regret Analysis in Gaussian Process Bandits: Optimality for Noiseless Reward, RKHS norm, and Non-Stationary Variance.

Shogo Iwazaki, Shion Takeno

Browse the full ICML paper archive.