# Building high accuracy emulators for scientific simulations with deep neural architecture search

arXiv: https://arxiv.org/abs/2001.08055 · HTML: https://mlsanity.com/paper/2001.08055/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Jan 2020.

## Built on by (most cited first, 30 of 36)

- [Finite Basis Physics-Informed Neural Networks (FBPINNs): a scalable domain decomposition approach for solving differential equations](https://mlsanity.com/paper/2107.07871/) (2021)
- [Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems](https://mlsanity.com/paper/2104.13386/) (2021)
- [A posteriori learning for quasi-geostrophic turbulence parametrization](https://mlsanity.com/paper/2204.03911/) (2022)
- [Fourier Neural Operators for Arbitrary Resolution Climate Data Downscaling](https://mlsanity.com/paper/2305.14452/) (2023)
- [DiffESM: Conditional Emulation of Temperature and Precipitation in Earth System Models with 3D Diffusion Models](https://mlsanity.com/paper/2409.11601/) (2024)
- [Can denoising diffusion probabilistic models generate realistic astrophysical fields?](https://mlsanity.com/paper/2211.12444/) (2022)
- [A Spatio-Temporal Neural Network Forecasting Approach for Emulation of Firefront Models](https://mlsanity.com/paper/2206.08523/) (2022)
- [NeurIPS 2024 ML4CFD Competition: Harnessing Machine Learning for Computational Fluid Dynamics in Airfoil Design](https://mlsanity.com/paper/2407.01641/) (2024)
- [Optimized and autonomous machine learning framework for characterizing pores, particles, grains and grain boundaries in microstructural images](https://mlsanity.com/paper/2101.06474/) (2021)
- [Few-shot Generation of Personalized Neural Surrogates for Cardiac Simulation via Bayesian Meta-Learning](https://mlsanity.com/paper/2210.02967/) (2022)
- [Data-driven Science and Machine Learning Methods in Laser-Plasma Physics](https://mlsanity.com/paper/2212.00026/) (2022)
- [Reduced Simulations for High-Energy Physics, a Middle Ground for Data-Driven Physics Research](https://mlsanity.com/paper/2309.03780/) (2023)
- [Feasibility Study on Active Learning of Smart Surrogates for Scientific Simulations](https://mlsanity.com/paper/2407.07674/) (2024)
- [High Throughput Training of Deep Surrogates from Large Ensemble Runs](https://mlsanity.com/paper/2309.16743/) (2023)
- [ML4PhySim : Machine Learning for Physical Simulations Challenge (The airfoil design)](https://mlsanity.com/paper/2403.01623/) (2024)
- [Quantum advantage for learning shallow neural networks with natural data distributions](https://mlsanity.com/paper/2503.20879/) (2025)
- [Opportunistic Emulation of Computationally Expensive Simulations via Deep Learning](https://mlsanity.com/paper/2108.11057/) (2021)
- [AI-coupled HPC Workflows](https://mlsanity.com/paper/2208.11745/) (2022)
- [On the Compatibility between Neural Networks and Partial Differential Equations for Physics-informed Learning](https://mlsanity.com/paper/2212.00270/) (2022)
- [Towards fast machine-learning-assisted Bayesian posterior inference of microseismic event location and source mechanism](https://mlsanity.com/paper/2101.04724/) (2021)
- [Self-learning Emulators and Eigenvector Continuation](https://mlsanity.com/paper/2107.13449/) (2021)
- [Learning Transport Processes with Machine Intelligence](https://mlsanity.com/paper/2109.13096/) (2021)
- [Exploring Randomly Wired Neural Networks for Climate Model Emulation](https://mlsanity.com/paper/2212.03369/) (2022)
- [Adapting U-Net for linear elastic stress estimation in polycrystal Zr microstructures](https://mlsanity.com/paper/2303.08541/) (2023)
- [A Neural Emulator for Uncertainty Estimation of Fire Propagation](https://mlsanity.com/paper/2305.06139/) (2023)
- [Machine Learning for Physical Simulation Challenge Results and Retrospective Analysis: Power Grid Use Case](https://mlsanity.com/paper/2505.01156/) (2025)
- [NeurIPS 2024 ML4CFD Competition: Results and Retrospective Analysis](https://mlsanity.com/paper/2506.08516/) (2025)
- [Designing Machine Learning Surrogates using Outputs of Molecular Dynamics Simulations as Soft Labels](https://mlsanity.com/paper/2110.14714/) (2021)
- [Reducing the Long Tail Losses in Scientific Emulations with Active Learning](https://mlsanity.com/paper/2111.08498/) (2021)
- [An Emulation Framework for Fire Front Spread](https://mlsanity.com/paper/2203.12160/) (2022)

## Abstract (arXiv)

Computer simulations are invaluable tools for scientific discovery. However, accurate simulations are often slow to execute, which limits their applicability to extensive parameter exploration, large-scale data analysis, and uncertainty quantification. A promising route to accelerate simulations by building fast emulators with machine learning requires large training datasets, which can be prohibitively expensive to obtain with slow simulations. Here we present a method based on neural architecture search to build accurate emulators even with a limited number of training data. The method successfully accelerates simulations by up to 2 billion times in 10 scientific cases including astrophysics, climate science, biogeochemistry, high energy density physics, fusion energy, and seismology, using the same super-architecture, algorithm, and hyperparameters. Our approach also inherently provides emulator uncertainty estimation, adding further confidence in their use. We anticipate this work will accelerate research involving expensive simulations, allow more extensive parameters exploration, and enable new, previously unfeasible computational discovery.