pi-Net is a hybrid physics-informed model that uses an encoder plus differentiable physics head for data-scarce multi-physics problems, demonstrated on a real-world Gas Metal Arc Welding case study.

#MachineLearning #AI #PhysicsInformedML #DataScience
https://arxiv.org/abs/2610.16504

New TurbRec: unsup. physics-consistent full-scale turbulence super-res. Low-res inputs + equations, no high-res data. Multi-scale Fourier embeddings; physics fixes small structures. Beats PINNs on spectra & stats.
link.springer.com/article/10.1...
#Turbulence #SuperRes #PhysicsInformedML