2000 character limit reached
Better Algorithms for Stochastic Bandits with Adversarial Corruptions (1902.08647v2)
Published 22 Feb 2019 in cs.LG and stat.ML
Abstract: We study the stochastic multi-armed bandits problem in the presence of adversarial corruption. We present a new algorithm for this problem whose regret is nearly optimal, substantially improving upon previous work. Our algorithm is agnostic to the level of adversarial contamination and can tolerate a significant amount of corruption with virtually no degradation in performance.