Towards multi-fidelity scaling laws of neural surrogates for CFD simulations

Paul Setinek, Gianluca Galletti, Johannes Brandstetter

Track: Track 1: Original Research/Position/Education/Attention Track

Keywords: multi-fidelity, scaling laws, neural surrogates, dataset composition

TL;DR: We study empirical scaling laws for neural surrogates trained on multi-fidelity CFD datasets.

Abstract:

Scaling laws describe how model performance grows with data, parameters and compute.
While large datasets can usually be collected at relatively low cost in domains such as language or vision, scientific machine learning is often limited by the high expense of generating training data through numerical simulations.
However, by adjusting modeling assumptions and approximations, simulation fidelity can be traded for computational cost, an aspect absent in other domains.
We investigate this trade-off between fidelity and cost in neural surrogates using low- and high-fidelity Reynolds-Averaged Navier-Stokes (RANS) simulations.
Reformulating classical scaling laws, we decompose the dataset axis into compute budget and dataset composition.
Our experiments reveal compute-performance scaling behavior and exhibit budget-dependent optimal fidelity mixes for the given dataset configuration.
These findings provide the first study of empirical scaling laws for multi-fidelity neural surrogate datasets and offer practical considerations for compute-efficient dataset generation.

Submission Number: 169