Integration with an Adaptive Harmonic Mean Algorithm

TL;DR

This paper introduces the Adaptive Harmonic Mean Integration (AHMI) algorithm, which estimates integrals in high-dimensional spaces using samples from MCMC, providing both the estimate and uncertainty, with demonstrated effectiveness on test cases.

Contribution

The paper presents a novel adaptive harmonic mean algorithm for integral estimation that improves accuracy and uncertainty quantification in high-dimensional Bayesian inference.

Findings

Effective in high-dimensional integral estimation

Provides reliable uncertainty quantification

Performs well on multiple test cases

Abstract

Numerically estimating the integral of functions in high dimensional spaces is a non-trivial task. A oft-encountered example is the calculation of the marginal likelihood in Bayesian inference, in a context where a sampling algorithm such as a Markov Chain Monte Carlo provides samples of the function. We present an Adaptive Harmonic Mean Integration (AHMI) algorithm. Given samples drawn according to a probability distribution proportional to the function, the algorithm will estimate the integral of the function and the uncertainty of the estimate by applying a harmonic mean estimator to adaptively chosen regions of the parameter space. We describe the algorithm and its mathematical properties, and report the results using it on multiple test cases.