Pde-refiner: Achieving accurate long rollouts with neural pde solvers

P Lippe, B Veeling, P Perdikaris… - Advances in …, 2024 - proceedings.neurips.cc
Time-dependent partial differential equations (PDEs) are ubiquitous in science and
engineering. Recently, mostly due to the high computational cost of traditional solution …

Weak baselines and reporting biases lead to overoptimism in machine learning for fluid-related partial differential equations

N McGreivy, A Hakim - Nature Machine Intelligence, 2024 - nature.com
One of the most promising applications of machine learning in computational physics is to
accelerate the solution of partial differential equations (PDEs). The key objective of machine …

Solving inverse problems in physics by optimizing a discrete loss: Fast and accurate learning without neural networks

P Karnakov, S Litvinov, P Koumoutsakos - PNAS nexus, 2024 - academic.oup.com
In recent years, advances in computing hardware and computational methods have
prompted a wealth of activities for solving inverse problems in physics. These problems are …

Modeling accurate long rollouts with temporal neural PDE solvers

P Lippe, BS Veeling, P Perdikaris… - ICML Workshop on …, 2023 - openreview.net
Time-dependent partial differential equations (PDEs) are ubiquitous in science and
engineering. Recently, mostly due to the high computational cost of traditional solution …

[HTML][HTML] FPL-net: A deep learning framework for solving the nonlinear Fokker–Planck–Landau collision operator for anisotropic temperature relaxation

H Noh, J Lee, E Yoon - Journal of Computational Physics, 2025 - Elsevier
The nonlinear collision operator consumes a significant amount of computation time in
tokamak whole-volume modeling, and in current numerical methods, the computational time …

Spectral-Refiner: Fine-Tuning of Accurate Spatiotemporal Neural Operator for Turbulent Flows

S Cao, F Brarda, R Li, Y ** - arxiv preprint arxiv:2405.17211, 2024 - arxiv.org
Recent advancements in operator-type neural networks have shown promising results in
approximating the solutions of spatiotemporal Partial Differential Equations (PDEs) …

PAPM: A Physics-aware Proxy Model for Process Systems

P Liu, Z Hao, X Ren, H Yuan, J Ren, D Ni - arxiv preprint arxiv …, 2024 - arxiv.org
In the context of proxy modeling for process systems, traditional data-driven deep learning
approaches frequently encounter significant challenges, such as substantial training costs …

[PDF][PDF] Fast and Accurate Simulation of Deformable Solid Dynamics on Coarse Meshes

MK Venturelli, W Celes - Available at SSRN 4879043, 2024 - maxwell.vrac.puc-rio.br
Matheus Kerber Venturelli Fast and Accurate Simulation of Deformable Solid Dynamics on
Coarse Meshes Page 1 Matheus Kerber Venturelli Fast and Accurate Simulation of …

Differentiable Programming for Computational Plasma Physics

NB McGreivy - 2024 - search.proquest.com
Differentiable programming allows for derivatives of functions implemented via computer
code to be calculated automatically. These derivatives are calculated using automatic …