Inverse Dirichlet weighting enables reliable training of physics informed neural networks
German Research Centre for Artificial Intelligence · Center for Systems Biology Dresden · Max Planck Institute of Molecular Cell Biology and Genetics · Center for Scalable Data Analytics and Artificial Intelligence · Technische Universität Dresden · Helmholtz-Zentrum Dresden-Rossendorf · Center for Advanced Systems Understanding · Helmholtz Munich · LMU Klinikum · Flatiron Health (United States) · Flatiron Institute · Ludwig-Maximilians-Universität München · Physics of Life
- 2021-11-05
- 128
@article{maddu2021inverse,
title = {Inverse Dirichlet weighting enables reliable training of physics informed neural networks},
author = {Suryanarayana Maddu and Dominik Sturm and Christian L. Müller and Ivo F. Sbalzarini},
year = {2021},
journal = {Machine Learning Science and Technology},
eprint = {2107.00940},
archivePrefix = {arXiv},
doi = {10.1088/2632-2153/ac3712},
url = {https://arxiv.org/abs/2107.00940},
}