Convergence rates for optimised adaptive importance samplers

Adaptive importance samplers are adaptive Monte Carlo algorithms to estimate expectations with respect to some target distribution which textitadapt themselves to obtain better estimators over a sequence of iterations. Although it is straightforward to show that they have the same O(1/√N) convergence rate as standard importance samplers, where N is the number of Monte Carlo samples, the behaviour of adaptive importance samplers over the number of iterations has been left relatively unexplored. In this work, we investigate an adaptation strategy based on convex optimisation which leads to a class of adaptive importance samplers termed textitoptimised adaptive importance samplers (OAIS). These samplers rely on the iterative minimisation of the chi2-divergence between an exponential-family proposal and the target. The analysed algorithms are closely related to the class of adaptive importance samplers which minimise the variance of the weight function. We first prove non-asymptotic error bounds for the mean squared errors (MSEs) of these algorithms, which explicitly depend on the number of iterations and the number of samples together. The non-asymptotic bounds derived in this paper imply that when the target belongs to the exponential family, the L2 errors of the optimised samplers converge to the optimal rate of O(1/√N) and the rate of convergence in the number of iterations are explicitly provided. When the target does not belong to the exponential family, the rate of convergence is the same but the asymptotic L2 error increases by a factor √ρ^star > 1, where ρ^star - 1 is the minimum chi2-divergence between the target and an exponential-family proposal.