This repository contains the R interface to the
Julia package NeuralEstimators (see here).
The package facilitates the user-friendly development of neural point
estimators, which are neural networks that map data to a point summary
of the posterior distribution. These estimators are likelihood-free and
amortised, in the sense that, after an initial setup cost, inference
from observed data can be made in a fraction of the time required by
conventional approaches. It also facilitates the construction of neural
networks that approximate the likelihood-to-evidence ratio in an
amortised fashion, which allows for making inference based on the
likelihood function or the entire posterior distribution. The package
caters for any model for which simulation is feasible by allowing the
user to implicitly define their model via simulated data. See the vignette
to get started!
To install the package, please:
Julia (see here) and R
(see here).NeuralEstimators.
julia -e 'using Pkg; Pkg.add(url="https://github.com/msainsburydale/NeuralEstimators.jl")'.R interface to
NeuralEstimators.
devtools in R and then run
devtools::install_github("msainsburydale/NeuralEstimators").Note that if you wish to simulate training data “on-the-fly” using
R functions, you will also need to install the Julia
package RCall. Note also that one may compile the vignettes
during installation (which takes roughly 10 minutes) by adding the
argument build_vignettes = TRUE in the final command
above.
This software was developed as part of academic research. If you would like to support it, please star the repository. If you use it in your research or other activities, please also use the following citation.
@article{,
author = {Sainsbury-Dale, Matthew and Zammit-Mangion, Andrew and Huser, Raphaël},
title = {Likelihood-Free Parameter Estimation with Neural {B}ayes Estimators},
journal = {The American Statistician},
year = {2024},
volume = {78},
pages = {1--14},
doi = {10.1080/00031305.2023.2249522},
url = {https://doi.org/10.1080/00031305.2023.2249522}
}Likelihood-free parameter estimation with neural Bayes estimators [paper] [code]
Neural Bayes estimators for censored inference with peaks-over-threshold models [paper]
Neural Bayes estimators for irregular spatial data using graph neural networks [paper][code]
Modern extreme value statistics for Utopian extremes [paper]
Several other software packages have been developed to facilitate neural likelihood-free inference. These include:
A summary of the functionality in these packages is given in Zammit-Mangion et al. (2024, Section 6.1). Note that this list of related packages was created in July 2024; if you have software to add to this list, please contact the package maintainer.