Search Results for author: Andrea Bonfanti

Found 2 papers, 0 papers with code

The Challenges of the Nonlinear Regime for Physics-Informed Neural Networks

no code implementations6 Feb 2024 Andrea Bonfanti, Giuseppe Bruno, Cristina Cipriani

The Neural Tangent Kernel (NTK) viewpoint represents a valuable approach to examine the training dynamics of Physics-Informed Neural Networks (PINNs) in the infinite width limit.

Second-order methods

On the Generalization of PINNs outside the training domain and the Hyperparameters influencing it

no code implementations15 Feb 2023 Andrea Bonfanti, Roberto Santana, Marco Ellero, Babak Gholami

Physics-Informed Neural Networks (PINNs) are Neural Network architectures trained to emulate solutions of differential equations without the necessity of solution data.

Cannot find the paper you are looking for? You can Submit a new open access paper.