AiNed XS Europa funding for generative AI for realistic physics simulations

The National Growth Fund programme AiNed allows promising ideas and innovative and speculative initiatives in the artificial intelligence domain to be explored. The projects are designed in collaboration with at least one European collaborative partner organization.

With this funding of 80K euros, CWI will launch the following project in collaboration with the Imperial College London; I-X Centre for AI in Science and the Technical University of Munich (TUM); School of Computation, Information and Technology.

Entropy-consistent learning: harnessing the power of generative AI for realistic physics simulations

This project develops a physics-aware generative AI model capable of generating physics simulations. While breakthroughs like ChatGPT and DALL-E have revolutionized generative modelling, generating correct results for physics problems like weather prediction is a challenge: contrary to text and images, physics problems often lack "big data" for training and require that generated simulations are adhering to the laws of physics. This project proposes entropy-consistent learning. This enforces physical correctness in the generative process. The generative model, trained on simulation data, is subsequently fine-tuned with observation data, again using entropy-consistent techniques. The resulting AI model seamlessly integrates data-driven techniques with physics-based models.

About the National Growth Fund programme AiNed

This Call for proposals is part of the National Growth Fund programme AiNed. This programme promotes the development and application of artificial intelligence (AI) in Dutch businesses and governments, and was developed by the Netherlands AI Coalition. Bringing the Netherlands into the leading group of AI countries for prosperity and well-being. That is the joint goal of the Dutch AI Coalition and the AiNed programme. Through a collaboration of public and private parties taking essential steps in developing and applying AI in various sectors and for important economic and societal challenges. In 2021, the AiNed proposal received a grant of € 204.5 million from the National Growth Fund.

About Benjamin Sanderse

Sanderse is the group leader of CWI’s Scientific Computing group. His work focuses on development of numerical methods for uncertainty quantification, for tackling closure problems, for constructing reduced order models, with the overarching theme of using structure-preserving techniques and applying them to solve complex partial differential equations, for example occurring in fluid flow problems. He obtained his PhD degree cum laude (with honours) in 2013 from Eindhoven University of Technology. Before starting his PhD degree, he received his MSc degree in Aerospace Engineering at Delft University of Technology in 2008. In 2020 Sanderse was awarded a NWO vidi grant.

About Nikolaj Mücke

Nikolaj Mücke is postdoctoral researcher in CWI's Scientific Computing Group. His work focuses on generative AI for physics applications, surrogate modeling, data assimilation, and uncertainty quantification. He obtained his PhD degree from Utrecht University in 2024. Before starting his PhD degree, he received his MSc degree in Mathematical Modeling and Computation from the Technical University of Denmark in 2019.

Read more about all the awarded projects on NWO.nl

Clouds photo: Shutterstock/Issaro Prakalung