Robust adaptive importance sampling for normal random vectors

TL;DR

This paper introduces a robust, automatic importance sampling method for normal random vectors that improves variance reduction efficiency by replacing stochastic approximation with sample average approximation and deterministic optimization.

Contribution

It proposes a novel importance sampling tuning approach that is more robust and automatic than traditional stochastic approximation methods, using sample average approximation and deterministic optimization.

Findings

Significant variance reduction compared to crude Monte Carlo.

Computation time reduced by a factor of 3 to 15 for the same precision.

Proven convergence and central limit theorem for the estimator.

Abstract

Adaptive Monte Carlo methods are very efficient techniques designed to tune simulation estimators on-line. In this work, we present an alternative to stochastic approximation to tune the optimal change of measure in the context of importance sampling for normal random vectors. Unlike stochastic approximation, which requires very fine tuning in practice, we propose to use sample average approximation and deterministic optimization techniques to devise a robust and fully automatic variance reduction methodology. The same samples are used in the sample optimization of the importance sampling parameter and in the Monte Carlo computation of the expectation of interest with the optimal measure computed in the previous step. We prove that this highly dependent Monte Carlo estimator is convergent and satisfies a central limit theorem with the optimal limiting variance. Numerical experiments confirm the performance of this estimator: in comparison with the crude Monte Carlo method, the computation time needed to achieve a given precision is divided by a factor between 3 and 15.