An Optimal Algorithm for Linear Bandits

10/19/2011

∙

by   Nicolò Cesa-Bianchi, et al.

∙

0

∙

share

We provide the first algorithm for online bandit linear optimization whose regret after T rounds is of order sqrtTd ln N on any finite class X of N actions in d dimensions, and of order d*sqrtT (up to log factors) when X is infinite. These bounds are not improvable in general. The basic idea utilizes tools from convex geometry to construct what is essentially an optimal exploration basis. We also present an application to a model of linear bandits with expert advice. Interestingly, these results show that bandit linear optimization with expert advice in d dimensions is no more difficult (in terms of the achievable regret) than the online d-armed bandit problem with expert advice (where EXP4 is optimal).