Home | Research | Groups | Nils Thuerey

Research Group Nils Thuerey


Link to website at TUM

Nils Thuerey

Prof. Dr.

Principal Investigator

Physics-based Simulation

Nils Thuerey

is Professor for Physics-based Simulation at TU Munich.

He works in the field of computer graphics, with a particular emphasis on physics-based deep learning algorithm. One focus of his research targets the simulation of fluid phenomena, such as water and smoke. These simulations find applications as visual effects in computer generated worlds, but also in many fields of engineering. Examples of his work are novel algorithms to make simulations easier to control, to handle detailed surface tension effects, and to increase the amount of turbulent detail.

Team members @MCML

PhD Students

Link to website

Felix Köhler

Physics-based Simulation

Link to website

Mohammad Rashed

Physics-based Simulation

Recent News @MCML

Link to Can AI Help Solve Complex Physics Equations? Meet APEBench

11.02.2025

Can AI Help Solve Complex Physics Equations? Meet APEBench

Link to MCML Researchers With 28 Papers at NeurIPS 2024

05.12.2024

MCML Researchers With 28 Papers at NeurIPS 2024

Publications @MCML

2025


[3]
K. Bhatia, F. Köhler and N. Thuerey.
PRDP: Progressively Refined Differentiable Physics.
ICLR 2025 - 13th International Conference on Learning Representations. Singapore, Apr 24-28, 2025. To be published. Preprint available. arXiv
Abstract

The physics solvers employed for neural network training are primarily iterative, and hence, differentiating through them introduces a severe computational burden as iterations grow large. Inspired by works in bilevel optimization, we show that full accuracy of the network is achievable through physics significantly coarser than fully converged solvers. We propose Progressively Refined Differentiable Physics (PRDP), an approach that identifies the level of physics refinement sufficient for full training accuracy. By beginning with coarse physics, adaptively refining it during training, and stopping refinement at the level adequate for training, it enables significant compute savings without sacrificing network accuracy. Our focus is on differentiating iterative linear solvers for sparsely discretized differential operators, which are fundamental to scientific computing. PRDP is applicable to both unrolled and implicit differentiation. We validate its performance on a variety of learning scenarios involving differentiable physics solvers such as inverse problems, autoregressive neural emulators, and correction-based neural-hybrid solvers. In the challenging example of emulating the Navier-Stokes equations, we reduce training time by 62%.

MCML Authors
Link to website

Felix Köhler

Physics-based Simulation

Link to Profile Nils Thuerey

Nils Thuerey

Prof. Dr.

Physics-based Simulation


[2]
Y. Shehata, B. Holzschuh and N. Thuerey.
Improved Sampling Of Diffusion Models In Fluid Dynamics With Tweedie's Formula.
ICLR 2025 - 13th International Conference on Learning Representations. Singapore, Apr 24-28, 2025. To be published. Preprint available. URL
Abstract

State-of-the-art Denoising Diffusion Probabilistic Models (DDPMs) rely on an expensive sampling process with a large Number of Function Evaluations (NFEs) to provide high-fidelity predictions. This computational bottleneck renders diffusion models less appealing as surrogates for the spatio-temporal prediction of physics-based problems with long rollout horizons. We propose Truncated Sampling Models, enabling single-step and few-step sampling with elevated fidelity by simple truncation of the diffusion process, reducing the gap between DDPMs and deterministic single-step approaches. We also introduce a novel approach, Iterative Refinement, to sample pre-trained DDPMs by reformulating the generative process as a refinement process with few sampling steps. Both proposed methods enable significant improvements in accuracy compared to DDPMs, DDIMs, and EDMs with NFEs 10 on a diverse set of experiments, including incompressible and compressible turbulent flow and airfoil flow uncertainty simulations. Our proposed methods provide stable predictions for long rollout horizons in time-dependent problems and are able to learn all modes of the data distribution in steady-state problems with high uncertainty.

MCML Authors
Link to Profile Nils Thuerey

Nils Thuerey

Prof. Dr.

Physics-based Simulation


2024


[1]
F. Köhler, S. Niedermayr, R. Westermann and N. Thuerey.
APEBench: A Benchmark for Autoregressive Neural Emulators of PDEs.
NeurIPS 2024 - 38th Conference on Neural Information Processing Systems. Vancouver, Canada, Dec 10-15, 2024. URL GitHub
Abstract

We introduce the Autoregressive PDE Emulator Benchmark (APEBench), a comprehensive benchmark suite to evaluate autoregressive neural emulators for solving partial differential equations. APEBench is based on JAX and provides a seamlessly integrated differentiable simulation framework employing efficient pseudo-spectral methods, enabling 46 distinct PDEs across 1D, 2D, and 3D. Facilitating systematic analysis and comparison of learned emulators, we propose a novel taxonomy for unrolled training and introduce a unique identifier for PDE dynamics that directly relates to the stability criteria of classical numerical methods. APEBench enables the evaluation of diverse neural architectures, and unlike existing benchmarks, its tight integration of the solver enables support for differentiable physics training and neural-hybrid emulators. Moreover, APEBench emphasizes rollout metrics to understand temporal generalization, providing insights into the long-term behavior of emulating PDE dynamics. In several experiments, we highlight the similarities between neural emulators and numerical simulators.

MCML Authors
Link to website

Felix Köhler

Physics-based Simulation

Link to Profile Rüdiger Westermann

Rüdiger Westermann

Prof. Dr.

Computer Graphics & Visualization

Link to Profile Nils Thuerey

Nils Thuerey

Prof. Dr.

Physics-based Simulation