This large-scale competition was met with great success with 237 participants and over 650 submissions!
In July, SystemX launched a NeurIPS 2024 competition titled “Harnessing Machine Learning for Computational Fluid Dynamics in Airfoil Design.” This competition, hosted on Codabench, focused on simulating airfoil design using the AirfRANS dataset and the LIPS platform provided by IRT SystemX. For the occasion, the institute partnered with Ansys, NVIDIA, Sorbonne University, Inria, and Criteo.
IRT SystemX announced the winning teams during the international NeurIPS conference:
- 1st prize: Fabien Casenave, Abbas Kabalan – Safran Tech, Digital Sciences & Technologies Department – École des Ponts ParisTech: MMGP: a Mesh Morphing Gaussian Process-based machine learning method for regression of physical problems under non-parameterized geometrical variability
- 2nd prize: Yuxin Ma Ant Research: Offset-based Graph Convolution for Mesh Graph
- 3rd prize: Giovanni Catalani, Jean Fesquet, Enrico Foglia, Xavier Bertrand, Frédéric Tost, Michaël Bauerheim, Josselin Morlier – Airbus (Toulouse, France) – ISAE-SUPAERO: MARIO: Multiscale Aerodynamic Resolution Invariant Operator.
- Student prize: Jacob Helwig, Xuan Zhang, Haiyang Yu, Shuiwang Ji – Texas A&M University : A Geometry-Aware Message Passing Neural Network for Modeling Aerodynamics over Airfoils
Congratulations to all!
