Approximate Bayesian Computation with Domain Expert in the Loop

TL;DR

This paper introduces an active learning approach for selecting summary statistics in Approximate Bayesian Computation, involving domain experts to improve inference, especially under model misspecification and limited simulation budgets.

Contribution

It presents a novel active learning method that reduces expert effort in ABC statistic selection and handles misspecified models effectively.

Findings

Better posterior estimates with limited simulations

Handles model misspecification effectively

Reduces domain expert workload

Abstract

Approximate Bayesian computation (ABC) is a popular likelihood-free inference method for models with intractable likelihood functions. As ABC methods usually rely on comparing summary statistics of observed and simulated data, the choice of the statistics is crucial. This choice involves a trade-off between loss of information and dimensionality reduction, and is often determined based on domain knowledge. However, handcrafting and selecting suitable statistics is a laborious task involving multiple trial-and-error steps. In this work, we introduce an active learning method for ABC statistics selection which reduces the domain expert's work considerably. By involving the experts, we are able to handle misspecified models, unlike the existing dimension reduction methods. Moreover, empirical results show better posterior estimates than with existing methods, when the simulation budget is limited.