On the Minimax Regret for Contextual Linear Bandits and Multi-Armed Bandits with Expert Advice

Open in new window