A review of Approximate Bayesian Computation methods via density estimation: inference for simulator-models

TL;DR

This paper reviews Approximate Bayesian Computation methods using density estimation, highlighting recent algorithms, their advantages, limitations, and potential future directions involving machine learning to improve scalability.

Contribution

It provides a comprehensive overview of ABC methods via density estimation, connecting traditional and recent machine learning approaches, and discusses future research directions.

Findings

Recent algorithms improve ABC efficiency

Machine learning offers scalable solutions for high-dimensional ABC

Advantages and limitations of parametric models are analyzed

Abstract

This paper provides a review of Approximate Bayesian Computation (ABC) methods for carrying out Bayesian posterior inference, through the lens of density estimation. We describe several recent algorithms and make connection with traditional approaches. We show advantages and limitations of models based on parametric approaches and we then draw attention to developments in machine learning, which we believe have the potential to make ABC scalable to higher dimensions and may be the future direction for research in this area.