Bandit Algorithms

171

familiar with information theory could skim this chapter. The final three chapters
are devoted to applying information theory to prove lower bounds on the regret
for both stochastic and adversarial bandits.