PINNs (Physics-informed Neural Networks)

A implementation of the Physics-informed Neural Networks (PINNs) using PyTorch was amazingly done by jayroxis/PINNs.

In this repository, solving WAVE EQUATION using PINN is replicated, following some experiment setup and procedures mentioned in the article:

@Article{app10175917, AUTHOR = {Guo, Yanan and Cao, Xiaoqun and Liu, Bainian and Gao, Mei}, TITLE = {Solving Partial Differential Equations Using Deep Learning and Physical Constraints}, DOI = {10.3390/app10175917}}

Credit

Original Work: Maziar Raissi, Paris Perdikaris, and George Em Karniadakis

Github Repo : https://github.com/maziarraissi/PINNs

Link: https://github.com/maziarraissi/PINNs/tree/master/appendix/continuous_time_identification%20(Burgers)

@article{raissi2017physicsI, title={Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations}, author={Raissi, Maziar and Perdikaris, Paris and Karniadakis, George Em}, journal={arXiv preprint arXiv:1711.10561}, year={2017} }

@article{raissi2017physicsII, title={Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations}, author={Raissi, Maziar and Perdikaris, Paris and Karniadakis, George Em}, journal={arXiv preprint arXiv:1711.10566}, year={2017} }