Improved Regret Bounds for Bandit Combinatorial Optimization

Open in new window