Bounded Regret for Finite-Armed Structured Bandits

Lattimore, Tor, Munos, Remi

Neural Information Processing Systems 

We study a new type of K-armed bandit problem where the expected return of one arm may depend on the returns of other arms. We present a new algorithm for this general class of problems and show that under certain circumstances it is possible to achieve finite expected cumulative regret. We also give problem-dependent lower bounds on the cumulative regret showing that at least in special cases the new algorithm is nearly optimal. Papers published at the Neural Information Processing Systems Conference.