no code implementations • 22 Oct 2018 • Alon Resler, Yishay Mansour
We present and study models of adversarial online learning where the feedback observed by the learner is noisy, and the feedback is either full information feedback or bandit feedback.
