Benchmarking Autoregressive Conditional Diffusion Models for Turbulent Flow Simulation

1 code implementation • 4 Sep 2023 • Georg Kohl, Li-Wei Chen, Nils Thuerey

We find that even simple diffusion-based approaches can outperform multiple established flow prediction methods in terms of accuracy and temporal stability, while being on par with state-of-the-art stabilization techniques like unrolling at training time.

Benchmarking

Learning Similarity Metrics for Volumetric Simulations with Multiscale CNNs

1 code implementation • 8 Feb 2022 • Georg Kohl, Li-Wei Chen, Nils Thuerey

Simulations that produce three-dimensional data are ubiquitous in science, ranging from fluid flows to plasma physics.

Learning Similarity Metrics for Numerical Simulations

1 code implementation • ICML 2020 • Georg Kohl, Kiwon Um, Nils Thuerey

We propose a neural network-based approach that computes a stable and generalizing metric (LSiM) to compare data from a variety of numerical simulation sources.