ActiveBayesOpInf

This repository contains the source code for the numerical experiments in the paper

Active learning for data-driven reduced models of parametric differential systems with Bayesian operator inference

by
Shane A. McQuarrie (Brigham Young University),
Mengwu Guo (Lund University), and
Anirban Chaudhuri (The University of Texas at Austin).

BibTex

@misc{mcquarrie2025activebayesopinf,
    title = {Active learning for data-driven reduced models of parametric differential systems with {B}ayesian operator inference},
    author = {Shane A. McQuarrie and Mengwu Guo and Anirban Chaudhuri},
    year = {2025},
    eprint = {2601.00038},
    archivePrefix = {arXiv},
}

Installation

This repository uses the standard Python scientific stack (NumPy, SciPy, Scikit-Learn, etc.) and the opinf package. We recommend installing the required packages in a new conda environment.

$ conda deactivate
$ conda create -n activebayesopinf python=3.13
$ conda activate activebayesopinf
(activebayesopinf) $ pip install -r requirements.txt

The files in the codebase/ folder are used in both experiments. Use make link in the terminal to set up links to these files.

Reproducing Numerical Results

The numerical experiments detailed in the paper are contained in separate folders.

• Heat equation with nonlinear reaction: heat/
• 2D viscous Burgers' equation: burgers/

To reproduce the figures in the paper, navigate to the directory and run experiments.sh.

cd heat/
./experiments.sh