Guardado en:
Detalles Bibliográficos
Autores principales: Bonfanti, Andrea, Medina, Ismael, List, Roman, Staeves, Björn, Santana, Roberto, Ellero, Marco
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2510.21262
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909867518722048
author Bonfanti, Andrea
Medina, Ismael
List, Roman
Staeves, Björn
Santana, Roberto
Ellero, Marco
author_facet Bonfanti, Andrea
Medina, Ismael
List, Roman
Staeves, Björn
Santana, Roberto
Ellero, Marco
contents Recent advances in Scientific Machine Learning have shown that second-order methods can enhance the training of Physics-Informed Neural Networks (PINNs), making them a suitable alternative to traditional numerical methods for Partial Differential Equations (PDEs). However, second-order methods induce large memory requirements, making them scale poorly with the model size. In this paper, we define a local Mixture of Experts (MoE) combining the parameter-efficiency of ensemble models and sparse coding to enable the use of second-order training. Our model -- \textsc{PINN Balls} -- also features a fully learnable domain decomposition structure, achieved through the use of Adversarial Adaptive Sampling (AAS), which adapts the DD to the PDE and its domain. \textsc{PINN Balls} achieves better accuracy than the state-of-the-art in scientific machine learning, while maintaining invaluable scalability properties and drawing from a sound theoretical background.
format Preprint
id arxiv_https___arxiv_org_abs_2510_21262
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PINN Balls: Scaling Second-Order Methods for PINNs with Domain Decomposition and Adaptive Sampling
Bonfanti, Andrea
Medina, Ismael
List, Roman
Staeves, Björn
Santana, Roberto
Ellero, Marco
Machine Learning
Recent advances in Scientific Machine Learning have shown that second-order methods can enhance the training of Physics-Informed Neural Networks (PINNs), making them a suitable alternative to traditional numerical methods for Partial Differential Equations (PDEs). However, second-order methods induce large memory requirements, making them scale poorly with the model size. In this paper, we define a local Mixture of Experts (MoE) combining the parameter-efficiency of ensemble models and sparse coding to enable the use of second-order training. Our model -- \textsc{PINN Balls} -- also features a fully learnable domain decomposition structure, achieved through the use of Adversarial Adaptive Sampling (AAS), which adapts the DD to the PDE and its domain. \textsc{PINN Balls} achieves better accuracy than the state-of-the-art in scientific machine learning, while maintaining invaluable scalability properties and drawing from a sound theoretical background.
title PINN Balls: Scaling Second-Order Methods for PINNs with Domain Decomposition and Adaptive Sampling
topic Machine Learning
url https://arxiv.org/abs/2510.21262