Finite-TimeRegretofThompsonSampling AlgorithmsforExponentialFamilyMulti-Armed Bandits
–Neural Information Processing Systems
Weprovideatightregretanalysis forExpTS, whichsimultaneously yields both the finite-timeregret bound as well as the asymptotic regret bound.
Neural Information Processing Systems
Feb-13-2026, 00:54:47 GMT