↓Skip to main content

🏢 Caltech

Pretraining Codomain Attention Neural Operators for Solving Multiphysics PDEs
·2558 words·13 mins· loading · loading
Machine Learning Few-Shot Learning 🏢 Caltech
CoDA-NO, a novel neural operator, revolutionizes multiphysics PDE solving via codomain tokenization, enabling efficient self-supervised pretraining and few-shot learning for superior generalization.