ANaGRAM: A Natural Gradient Relative to Adapted Model for efficient PINNs learning
Date:
Presentation of ANaGRAM at ICLR 2025. ANaGRAM is a theoretically founded and algorithmically efficient natural gradient method for training Physics-Informed Neural Networks (PINNs), connecting natural gradient descent to the Neural Tangent Kernel and the Green function of the PDE operator.
| Slides | Poster | ICLR page | Proceedings |