Probabilistic graph networks for learning physics simulations

DOI10.1016/J.JCP.2024.113137MaRDI QIDQ6572165zbMATH OpenFDO

Authors Sakthi Kumar Arul Prakash, Conrad Tucker

Publication date 15 July 2024

Published in Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.JCP.2024.113137

zbMATH Keywords

graph networksnormalizing flowsphysics simulations

Mathematics Subject Classification ID

Nonparametric inference (62Gxx) Inference from stochastic processes (62Mxx) Artificial intelligence (68Txx)

Recommendations

• Learning generative neural networks with physics knowledge
• GINNs: graph-informed neural networks for multiscale physics
• Predicting dynamic responses of continuous deformable bodies: a graph-based learning approach
• Physics-informed \textsc{MeshGraphNets} (PI-MGNs): neural finite element solvers for non-stationary and nonlinear simulations on arbitrary meshes
• Differentiability in unrolled training of neural physics simulators on transient dynamics

Cites work

• Converting high-dimensional regression to high-dimensional conditional density estimation
• GINNs: graph-informed neural networks for multiscale physics
• Learning to predict the cosmological structure formation
• Physics-informed multi-LSTM networks for metamodeling of nonlinear structures
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Transformers for modeling physical systems

This page was built for publication: Probabilistic graph networks for learning physics simulations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6572165)