UltraNest -- a robust, general purpose Bayesian inference engine

TL;DR

UltraNest is a versatile Bayesian inference tool designed for accurate, fast parameter estimation and model comparison, especially effective for complex, multi-modal, and computationally intensive models, supporting multiple programming languages and parallel computing.

Contribution

It introduces a robust, multi-language compatible Bayesian inference engine optimized for correctness and speed, with features like parallelization and run resumption.

Findings

Effective handling of multi-modal and non-Gaussian spaces

Supports various programming languages for model specification

Enables parallel computing and resuming incomplete runs

Abstract

UltraNest is a general-purpose Bayesian inference package for parameter estimation and model comparison. It allows fitting arbitrary models specified as likelihood functions written in Python, C, C++, Fortran, Julia or R. With a focus on correctness and speed (in that order), UltraNest is especially useful for multi-modal or non-Gaussian parameter spaces, computational expensive models, in robust pipelines. Parallelisation to computing clusters and resuming incomplete runs is available.