AutoEmulate: A PyTorch tool for end-to-end emulation workflows

Computational simulations lie at the heart of modern science and engineering, but they are often slow and computationally costly.A common solution is to use emulators: fast, cheap models trained to approximate the simulator.However, constructing these requires substantial expertise.AutoEmulate (Stoffel et al., 2025) is a Python package for end-to-end emulation workflows that automates much of the machine learning pipeline, making it easy to replace simulations with fast, accurate emulators.AutoEmulate has now been fully refactored to use PyTorch as the primary and default backend, enabling GPU acceleration, automatic differentiation, and seamless integration with the broader PyTorch ecosystem.The toolkit has also been extended with easy-to-use interfaces for common emulation tasks, including model calibration (determining which input values are most likely to have generated real-world observations) and active learning (where simulations are chosen to improve emulator performance at minimal computational cost).Together these updates make AutoEmulate uniquely suited to running performant end-to-end emulation workflows.

Authors

Institutions

Publication Details

Journal
The Journal of Open Source Software
Published
2026-08-26
DOI
https://doi.org/10.21105/joss.10087
Primary Topic
Scientific Computing and Data Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

AutoEmulate: A PyTorch tool for end-to-end emulation workflows

Steven Niederer, Sophie Arana, Jason D. McEwen, Edward Chalstrey et al.
The Journal of Open Source Software
Scientific Computing and Data Management
article

AutoEmulate: A PyTorch tool for end-to-end emulation workflows

Steven Niederer, Sophie Arana, Jason D. McEwen, Edward Chalstrey, Radka Jersakova, Christopher Iliffe Sprague, E. G. Daub, Camila Rangel Smith, Maximilian Balmus, Paolo Conti, Bryan M. Li, Martin A. Stoffel, Kalle Westerling, Marjan Famili, Sam F. Greenbury, A. Duncan, Edwin Brown
article en

Abstract

Computational simulations lie at the heart of modern science and engineering, but they are often slow and computationally costly.A common solution is to use emulators: fast, cheap models trained to approximate the simulator.However, constructing these requires substantial expertise.AutoEmulate (Stoffel et al., 2025) is a Python package for end-to-end emulation workflows that automates much of the machine learning pipeline, making it easy to replace simulations with fast, accurate emulators.AutoEmulate has now been fully refactored to use PyTorch as the primary and default backend, enabling GPU acceleration, automatic differentiation, and seamless integration with the broader PyTorch ecosystem.The toolkit has also been extended with easy-to-use interfaces for common emulation tasks, including model calibration (determining which input values are most likely to have generated real-world observations) and active learning (where simulations are chosen to improve emulator performance at minimal computational cost).Together these updates make AutoEmulate uniquely suited to running performant end-to-end emulation workflows.

The Journal of Open Source SoftwareVol. 11(124)
The Alan Turing Institute (GB), University College London (GB), Imperial College London (GB), University of Sheffield (GB), University of Edinburgh (GB)
Life below water
Openalex Percentile: Top 21%
Scientific Computing and Data Management
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.