On Estimating the Gradient of the Expected Information Gain in Bayesian Experimental Design

TL;DR

This paper develops and compares two novel methods for efficiently estimating the gradient of the expected information gain in Bayesian experimental design, enabling more effective optimization of experimental conditions.

Contribution

Introduces two new gradient estimation methods, UEEG-MCMC and BEEG-AP, for Bayesian experimental design, with theoretical analysis and superior numerical performance.

Findings

UEEG-MCMC is robust regardless of the EIG value.

BEEG-AP is more efficient for small EIG values.

Both methods outperform existing benchmarks in experiments.

Abstract

Bayesian Experimental Design (BED), which aims to find the optimal experimental conditions for Bayesian inference, is usually posed as to optimize the expected information gain (EIG). The gradient information is often needed for efficient EIG optimization, and as a result the ability to estimate the gradient of EIG is essential for BED problems. The primary goal of this work is to develop methods for estimating the gradient of EIG, which, combined with the stochastic gradient descent algorithms, result in efficient optimization of EIG. Specifically, we first introduce a posterior expected representation of the EIG gradient with respect to the design variables. Based on this, we propose two methods for estimating the EIG gradient, UEEG-MCMC that leverages posterior samples generated through Markov Chain Monte Carlo (MCMC) to estimate the EIG gradient, and BEEG-AP that focuses on achieving high simulation efficiency by repeatedly using parameter samples. Theoretical analysis and numerical studies illustrate that UEEG-MCMC is robust agains the actual EIG value, while BEEG-AP is more efficient when the EIG value to be optimized is small. Moreover, both methods show superior performance compared to several popular benchmarks in our numerical experiments.