Unifying Summary Statistic Selection for Approximate Bayesian Computation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hoffmann, Till, Onnela, Jukka-Pekka
Format: Preprint
Veröffentlicht: 2022
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908669766008832
author Hoffmann, Till
Onnela, Jukka-Pekka
author_facet Hoffmann, Till
Onnela, Jukka-Pekka
contents Extracting low-dimensional summary statistics from large datasets is essential for efficient (likelihood-free) inference. We characterize three different classes of summaries and demonstrate their importance for correctly analyzing dimensionality reduction algorithms. We demonstrate that minimizing the expected posterior entropy (EPE) under the prior predictive distribution of the model provides a unifying principle that subsumes many existing methods; they are shown to be equivalent to, or special or limiting cases of, minimizing the EPE. We offer a unifying framework for obtaining informative summaries and propose a practical method using conditional density estimation to learn high-fidelity summaries automatically. We evaluate this approach on diverse problems, including a challenging benchmark model with a multi-modal posterior, a population genetics model, and a dynamic network model of growing trees. The results show that EPE-minimizing summaries can lead to posterior inference that is competitive with, and in some cases superior to, dedicated likelihood-based approaches, providing a powerful and general tool for practitioners.
format Preprint
id arxiv_https___arxiv_org_abs_2206_02340
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Unifying Summary Statistic Selection for Approximate Bayesian Computation
Hoffmann, Till
Onnela, Jukka-Pekka
Methodology
Information Theory
Machine Learning
Extracting low-dimensional summary statistics from large datasets is essential for efficient (likelihood-free) inference. We characterize three different classes of summaries and demonstrate their importance for correctly analyzing dimensionality reduction algorithms. We demonstrate that minimizing the expected posterior entropy (EPE) under the prior predictive distribution of the model provides a unifying principle that subsumes many existing methods; they are shown to be equivalent to, or special or limiting cases of, minimizing the EPE. We offer a unifying framework for obtaining informative summaries and propose a practical method using conditional density estimation to learn high-fidelity summaries automatically. We evaluate this approach on diverse problems, including a challenging benchmark model with a multi-modal posterior, a population genetics model, and a dynamic network model of growing trees. The results show that EPE-minimizing summaries can lead to posterior inference that is competitive with, and in some cases superior to, dedicated likelihood-based approaches, providing a powerful and general tool for practitioners.
title Unifying Summary Statistic Selection for Approximate Bayesian Computation
topic Methodology
Information Theory
Machine Learning
url https://arxiv.org/abs/2206.02340