Title: Multidimensional monopolist pricing problem with uncertain valuations

Abstract: We consider the problem of revenue maximization by a monopolist selling a set of substitutable goods to an unit-demand buyer of uncertain type. The Bayesian variant of this problem has received much interest in the computer science and economics communities in the past years. Here we assume that no distribution of buyer's type is known. We consider the problem of finding robust solutions in the min-max regret sense. We give polynomial time algorithms for optimal robust pricing when uncertainty is modeled by intervals or discrete scenario sets.