Generative learning of the solution of parametric partial differential equations using guided diffusion models and virtual observations

H Gao, S Kaltenbach, P Koumoutsakos - Computer Methods in Applied …, 2025 - Elsevier
We introduce a generative learning framework to model high-dimensional parametric
systems using gradient guidance and virtual observations. We consider systems described …

Semi-supervised invertible neural operators for Bayesian inverse problems

S Kaltenbach, P Perdikaris… - Computational Mechanics, 2023 - Springer
Neural Operators offer a powerful, data-driven tool for solving parametric PDEs as they can
represent maps between infinite-dimensional function spaces. In this work, we employ …

STENCIL-NET for equation-free forecasting from data

S Maddu, D Sturm, BL Cheeseman, CL Müller… - Scientific reports, 2023 - nature.com
We present an artificial neural network architecture, termed STENCIL-NET, for equation-free
forecasting of spatiotemporal dynamics from data. STENCIL-NET works by learning a …

Fractional rheology-informed neural networks for data-driven identification of viscoelastic constitutive models

D Dabiri, M Saadat, D Mangal, S Jamali - Rheologica Acta, 2023 - Springer
Developing constitutive models that can describe a complex fluid's response to an applied
stimulus has been one of the critical pursuits of rheologists. The complexity of the models …

Personalized predictions of Glioblastoma infiltration: Mathematical models, Physics-Informed Neural Networks and multimodal scans

RZ Zhang, I Ezhov, M Balcerak, A Zhu, B Wiestler… - Medical Image …, 2025 - Elsevier
Predicting the infiltration of Glioblastoma (GBM) from medical MRI scans is crucial for
understanding tumor growth dynamics and designing personalized radiotherapy treatment …

Physics-informed neural networks for incompressible flows with moving boundaries

Y Zhu, W Kong, J Deng, X Bian - Physics of Fluids, 2024 - pubs.aip.org
Physics-informed neural networks (PINNs) employed in fluid mechanics deal primarily with
stationary boundaries. This hinders the capability to address a wide range of flow problems …

Bilo: Bilevel local operator learning for pde inverse problems

RZ Zhang, X Xie, JS Lowengrub - arXiv preprint arXiv:2404.17789, 2024 - arxiv.org
We propose a new neural network based method for solving inverse problems for partial
differential equations (PDEs) by formulating the PDE inverse problem as a bilevel …

Neural functional a posteriori error estimates

V Fanaskov, A Rudikov, I Oseledets - arXiv preprint arXiv:2402.05585, 2024 - arxiv.org
We propose a new loss function for supervised and physics-informed training of neural
networks and operators that incorporates a posteriori error estimate. More specifically …

Flow reconstruction by multiresolution optimization of a discrete loss with automatic differentiation

P Karnakov, S Litvinov, P Koumoutsakos - The European Physical Journal …, 2023 - Springer
We present a potent computational method for the solution of inverse problems in fluid
mechanics. We consider inverse problems formulated in terms of a deterministic loss …

Electromagnetic-thermal analysis with FDTD and physics-informed neural networks

S Qi, CD Sarris - IEEE Journal on Multiscale and Multiphysics …, 2023 - ieeexplore.ieee.org
This article presents the coupling of the finite-difference time-domain (FDTD) method for
electromagnetic field simulation, with a physics-informed neural network based solver for the …