Multilevel Monte Carlo estimation of expected information gains

Abstract

The expected information gain is an important quality criterion of Bayesian experimental designs, which measures how much the information entropy about uncertain quantity of interest $\theta$ is reduced on average by collecting relevant data $Y$. However, estimating the expected information gain has been considered computationally challenging since it is defined as a nested expectation with an outer expectation with respect to $Y$ and an inner expectation with respect to $\theta$. In fact, the standard, nested Monte Carlo method requires a total computational cost of $O({\epsilon }^{-3})$ to achieve a root-mean-square accuracy of $\epsilon$. In this paper we develop an efficient algorithm to estimate the expected information gain by applying a multilevel Monte Carlo (MLMC) method. To be precise, we introduce an antithetic MLMC estimator for the expected information gain and provide a sufficient condition on the data model under which the antithetic property of the MLMC estimator is well exploited such that optimal complexity of $O({\epsilon }^{-2})$ is achieved. Furthermore, we discuss how to incorporate importance sampling techniques within the MLMC estimator to avoid arithmetic underflow. Numerical experiments show the considerable computational cost savings compared to the nested Monte Carlo method for a simple test case and a more realistic pharmacokinetic model.

References in corpus (4)

• Simulation-based optimal Bayesian experimental design for nonlinear systems
• Antithetic multilevel Monte Carlo estimation for multi-dimensional SDEs without Lévy area simulation
• Fast Bayesian experimental design: Laplace-based importance sampling for the expected information gain
• Decision-making under uncertainty: using MLMC for efficient estimation of EVPPI

Cited by in corpus (9)

• Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs
• Multilevel Monte Carlo estimation of the expected value of sample information
• Multi-index ensemble Kalman filtering
• On the effective dimension and multilevel Monte Carlo
• The Roles of Low-Noise Stations, Arrays and Ocean-Bottom Seismometers in Monitoring UK Offshore Seismicity associated with Subsurface Storage of Carbon Dioxide
• A Multi-fidelity Estimator of the Expected Information Gain for Bayesian Optimal Experimental Design
• Optimal Experimental Design for Reliable Learning of History-Dependent Constitutive Laws
• Constructing unbiased gradient estimators with finite variance for conditional stochastic optimization
• An efficient estimation of nested expectations without conditional sampling