The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index.
Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
Abstract
The widespread use of neural networks across different scientific domains often involves constraining them to satisfy certain symmetries, conservation laws, or other domain knowledge. Such constraints are often imposed as soft penalties during model training and effectively act as domain-specific regularizers of the empirical risk loss. Physics-informed neural networks is an example of this philosophy in which the outputs of deep neural networks are constrained to approximately satisfy a given set of partial differential equations. In this work we review recent advances in scientific machine learning with a specific focus on the effectiveness of physics-informed neural networks in predicting outcomes of physical systems and discovering hidden physics from noisy data. We also identify and analyze a fundamental mode of failure of such approaches that is related to numerical stiffness leading to unbalanced back-propagated gradients during model training. To address this limitation we present a learning rate annealing algorithm that utilizes gradient statistics during model training to balance the interplay between different terms in composite loss functions. We also propose a novel neural network architecture that is more resilient to such gradient pathologies. Taken together, our developments provide new insights into the training of constrained neural networks and consistently improve the predictive accuracy of physics-informed neural networks by a factor of 50--100$\times$ across a range of problems in computational physics. All code and data accompanying this manuscript are publicly available at https://github.com/PredictiveIntelligenceLab/GradientPathologiesPINNs.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Raissi, M., Perdikaris, P., & Karniadakis, G. E. (2019) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.10.045 · Cited by 20,384
Self-citationProceedings of IEEE Conference on Computer Vision and Pattern Recognition
(1993) · Proceedings of IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.1993.341010 · Cited by 14,265
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,887
Advances in Neural Information Processing Systems 25
Zhang, Y., Sutton, C. A., Storkey, A., & Zoubin Ghahramani, A. (2012) · Cited by 12,653
Unsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235
Mastering the game of Go without human knowledge
Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., Hubert, T., Baker, L., Lai, M., Bolton, A., et al. (2017) · Nature · doi:10.1038/nature24270 · Cited by 9,084
Advances in Neural Information Processing Systems 14: Proceedings of the 2001 Conference
(2002) · The MIT Press eBooks · doi:10.7551/mitpress/1120.001.0001 · Cited by 8,890
Parikh, N., & Boyd, S. (2014) · Foundations and Trends® in Optimization · doi:10.1561/2400000003 · Cited by 4,102
On the Construction and Comparison of Difference Schemes
Strang, G. (1968) · SIAM Journal on Numerical Analysis · doi:10.1137/0705041 · Cited by 3,571
Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning
Alipanahi, B., Delong, A., Weirauch, M. T., & Frey, B. J. (2015) · Nature Biotechnology · doi:10.1038/nbt.3300 · Cited by 2,873
Relativistic Quantum Mechanics
Bjorken, J. D., Drell, S. D., & Mansfield, J. E. (1965) · Physics Today · doi:10.1063/1.3047288 · Cited by 2,836
Artificial neural networks for solving ordinary and partial differential equations
Lagaris, I. E., Likas, A., & Fotiadis, D. I. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/72.712178 · Cited by 2,412
DGM: A deep learning algorithm for solving partial differential equations
Sirignano, J., & Spiliopoulos, K. (2018) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.08.029 · Cited by 2,208
Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
Raissi, M., Yazdani, A., & Karniadakis, G. E. (2020) · Science · doi:10.1126/science.aaw4741 · Cited by 2,103
Searching for exotic particles in high-energy physics with deep learning
Baldi, P., Sadowski, P., & Whiteson, D. (2014) · Nature Communications · doi:10.1038/ncomms5308 · Cited by 1,164
Sun, L., Gao, H., Pan, S., & Wang, J.-X. (2020) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2019.112732 · Cited by 1,045
Zhu, Y., Zabaras, N., Koutsourelakis, P.-S., & Perdikaris, P. (2019) · Journal of Computational Physics · doi:10.1016/j.jcp.2019.05.024 · Cited by 1,030
Self-citationA hybrid neural network‐first principles approach to process modeling
Psichogios, D. C., & Ungar, L. H. (1992) · AIChE Journal · doi:10.1002/aic.690381003 · Cited by 982
Artificial Neural Networks for Solving Ordinary and Partial Differential Equations
Lagaris, I. E., Likas, A., & Fotiadis, D. I. (1998) · Cited by 944
Understanding deep convolutional networks
Mallat, S. (2016) · Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rsta.2015.0203 · Cited by 688
Kissas, G., Yang, Y., Hwuang, E., Witschey, W. R., Detre, J. A., & Perdikaris, P. (2020) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2019.112623 · Cited by 647
Self-citationTartakovsky, A. M., Marrero, C. O., Perdikaris, P., Tartakovsky, G. D., & Barajas‐Solano, D. (2020) · Water Resources Research · doi:10.1029/2019wr026731 · Cited by 515
Self-citationThe 2D lid-driven cavity problem revisited
Bruneau, C.-H., & Saad, M. (2006) · Computers & Fluids · doi:10.1016/j.compfluid.2004.12.004 · Cited by 465
Tripathy, R. K., & Bilionis, I. (2018) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.08.036 · Cited by 464
Adversarial uncertainty quantification in physics-informed neural networks
Yang, Y., & Perdikaris, P. (2019) · Journal of Computational Physics · doi:10.1016/j.jcp.2019.05.027 · Cited by 414
Self-citationModeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks
Geneva, N., & Zabaras, N. (2020) · Journal of Computational Physics · doi:10.1016/j.jcp.2019.109056 · Cited by 339
Shin, Y., Darbon, J., & Em Karniadakis, G. (2020) · Communications in Computational Physics · doi:10.4208/cicp.oa-2020-0193 · Cited by 284
Cited by
Physics-informed machine learning
Karniadakis, G. E., Kevrekidis, I. G., Lu, L., Perdikaris, P., Wang, S., & Yang, L. (2021) · Nature Reviews Physics · doi:10.1038/s42254-021-00314-5 · Cited by 8,718
Self-citationScientific Machine Learning Through Physics–Informed Neural Networks: Where we are and What’s Next
Cuomo, S., Di Cola, V. S., Giampaolo, F., Rozza, G., Raissi, M., & Piccialli, F. (2022) · Journal of Scientific Computing · doi:10.1007/s10915-022-01939-z · Cited by 2,784
Alzubaidi, L., Bai, J., Al-Sabaawi, A., Santamaría, J., Albahri, A. S., Al-dabbagh, B. S. N., Fadhel, M. A., Manoufali, M., Zhang, J., Al-Timemy, A. H., et al. (2023) · Journal of Big Data · doi:10.1186/s40537-023-00727-2 · Cited by 907
Integrating Scientific Knowledge with Machine Learning for Engineering and Environmental Systems
Willard, J., Jia, X., Xu, S., Steinbach, M., & Kumar, V. (2023) · ACM Computing Surveys · doi:10.1145/3514228 · Cited by 878
Wu, C., Zhu, M., Tan, Q., Kartha, Y., & Lu, L. (2023) · Computer Methods in Applied Mechanics and Engineering · doi:10.1016/j.cma.2022.115671 · Cited by 723
Metadata
- Authors
- Wang, S., Teng, Y., & Perdikaris, P.
- Venue
- SIAM Journal on Scientific Computing
- Published
- 2021-01
- Volume
- 43
- Issue
- 5
- Pages
- A3055-A3081
- ISSN
- 1064-8275 (ISSN-L)
- 1095-7197
- OpenAlex
- W3200673624
- Cited by
- 2,048
- References
- 27
- FWCI
- 71.06
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.