In recent years methods from optimal linear experimental design have been leveraged to obtain state of the art results for linear bandits. A design returned from an objective such as -optimal design is actually a probability distribution over a pool of potential measurement vectors. Consequently, one nuisance of the approach is the task of converting this continuous probability distribution into a discrete assignment of measurements. While sophisticated rounding techniques have been proposed, in dimensions they require to be at least , , or based on the sub-optimality of the solution. In this paper we are interested in settings where may be much less than , such as in experimental design in an RKHS where may be effectively infinite. In this work, we propose a rounding procedure that frees of any dependence on the dimension , while achieving nearly the same performance guarantees of existing rounding procedures. We evaluate the procedure against a baseline that projects the problem to a lower dimensional space and performs rounding which requires to just be at least a notion of the effective dimension. We also leverage our new approach in a new algorithm for kernelized bandits to obtain state of the art results for regret minimization and pure exploration. An advantage of our approach over existing UCB-like approaches is that our kernel bandit algorithms are also robust to model misspecification.
View on arXiv