Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
184 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

ST-PINN: A Self-Training Physics-Informed Neural Network for Partial Differential Equations (2306.09389v1)

Published 15 Jun 2023 in cs.LG, cs.AI, and physics.comp-ph

Abstract: Partial differential equations (PDEs) are an essential computational kernel in physics and engineering. With the advance of deep learning, physics-informed neural networks (PINNs), as a mesh-free method, have shown great potential for fast PDE solving in various applications. To address the issue of low accuracy and convergence problems of existing PINNs, we propose a self-training physics-informed neural network, ST-PINN. Specifically, ST-PINN introduces a pseudo label based self-learning algorithm during training. It employs governing equation as the pseudo-labeled evaluation index and selects the highest confidence examples from the sample points to attach the pseudo labels. To our best knowledge, we are the first to incorporate a self-training mechanism into physics-informed learning. We conduct experiments on five PDE problems in different fields and scenarios. The results demonstrate that the proposed method allows the network to learn more physical information and benefit convergence. The ST-PINN outperforms existing physics-informed neural network methods and improves the accuracy by a factor of 1.33x-2.54x. The code of ST-PINN is available at GitHub: https://github.com/junjun-yan/ST-PINN.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (32)
  1. R. Temam and A. Chorin, “Navier stokes equations: Theory and numerical analysis,” Journal of Applied Mechanics, vol. 2, no. 2, p. 456, 1984.
  2. M. Raissi, A. Yazdani, and G. E. Karniadakis, “Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations,” Science (New York, N.Y.), pp. 1026–1030, 2020.
  3. A. Jaun, J. Hedin, and T. Johnson, “Numerical methods for partial differential equations,” Swedish Netuniversity, 1999.
  4. G. E. Karniadakis, I. G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang, and L. Yang, “Physics-informed machine learning,” Nature Reviews Physics, vol. 3, no. 6, pp. 422–440, 2021.
  5. S. Cai, Z. Mao, Z. Wang, M. Yin, and G. E. Karniadakis, “Physics-informed neural networks (pinns) for fluid mechanics: A review,” Acta Mechanica Sinica, vol. 37, no. 12, pp. 1727–1738, 2021.
  6. O. Aguilar-Leal, R. Q. Fuentes-Aguilar, I. Chairez, A. García-González, and J. C. Huegel, “Distributed parameter system identification using finite element differential neural networks,” Applied Soft Computing, vol. 43, pp. 633–642, 2016.
  7. K. Patan and M. Patan, “Neural-network-based models ensemble for identification in distributed-parameter systems with application to elastic materials modeling,” in 2022 International Joint Conference on Neural Networks (IJCNN), 2022, pp. 01–08.
  8. M. Raissi, P. Perdikaris, and G. Karniadakis, “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations,” Journal of Computational Physics, vol. 378, pp. 686–707, 2019.
  9. L. Lu, X. Meng, Z. Mao, and G. E. Karniadakis, “Deepxde: A deep learning library for solving differential equations,” SIAM Rev., vol. 63, no. 1, pp. 208–228, 2021.
  10. T. D. Ryck and S. Mishra, “Generic bounds on the approximation error for physics-informed (and) operator learning,” in Advances in Neural Information Processing Systems, 2022.
  11. Y. Shin, J. Darbon, and G. E. Karniadakis, “On the convergence and generalization of physics informed neural networks,” arXiv e-prints, pp. arXiv–2004, 2020.
  12. S. Mishra and R. Molinaro, “Estimates on the generalization error of physics-informed neural networks for approximating a class of inverse problems for pdes,” IMA Journal of Numerical Analysis, vol. 42, no. 2, pp. 981–1022, 2022.
  13. S. Wang, X. Yu, and P. Perdikaris, “When and why pinns fail to train: A neural tangent kernel perspective,” Journal of Computational Physics, vol. 449, p. 110768, 2022.
  14. X. Chen, Y. Yuan, G. Zeng, and J. Wang, “Semi-supervised semantic segmentation with cross pseudo supervision,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 2613–2622.
  15. J. E. Van Engelen and H. H. Hoos, “A survey on semi-supervised learning,” Machine learning, vol. 109, no. 2, pp. 373–440, 2020.
  16. L. Xu, C. Hu, and K. Mei, “Semi-supervised regression with manifold: A bayesian deep kernel learning approach,” Neurocomputing, vol. 497, pp. 76–85, 2022.
  17. B. Zoph, G. Ghiasi, T.-Y. Lin, Y. Cui, H. Liu, E. D. Cubuk, and Q. Le, “Rethinking pre-training and self-training,” in Advances in Neural Information Processing Systems, vol. 33.   Curran Associates, Inc., 2020, pp. 3833–3845.
  18. P. Cascante-Bonilla, F. Tan, Y. Qi, and V. Ordonez, “Curriculum labeling: Revisiting pseudo-labeling for semi-supervised learning,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 8, 2021, pp. 6912–6920.
  19. A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Commun. ACM, vol. 60, no. 6, pp. 84–90, 2017.
  20. X. Chen, C. Gong, J. Liu, Y. Pang, L. Deng, L. Chi, and K. Li, “A novel neural network approach for airfoil mesh quality evaluation,” Journal of Parallel and Distributed Computing, vol. 164, pp. 123–132, 2022.
  21. I. E. Lagaris, A. Likas, and D. I. Fotiadis, “Artificial neural networks for solving ordinary and partial differential equations,” IEEE transactions on neural networks, vol. 9, no. 5, pp. 987–1000, 1998.
  22. S. Seo, C. Meng, S. Rambhatla, and Y. Liu, “Physics-aware spatiotemporal modules with auxiliary tasks for meta-learning,” in Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21, 8 2021, pp. 2943–2949.
  23. F. Sun, Y. Liu, and H. Sun, “Physics-informed spline learning for nonlinear dynamics discovery,” in Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21, 8 2021, pp. 2054–2061.
  24. A. D. Jagtap, E. Kharazmi, and G. E. Karniadakis, “Conservative physics-informed neural networks on discrete domains for conservation laws: Applications to forward and inverse problems,” Computer Methods in Applied Mechanics and Engineering, vol. 365, p. 113028, 2020.
  25. A. D. Jagtap and G. E. Karniadakis, “Extended physics-informed neural networks (xpinns): A generalized space-time domain decomposition based deep learning framework for nonlinear partial differential equations,” in Proceedings of the AAAI 2021 Spring Symposium on Combining Artificial Intelligence and Machine Learning with Physical Sciences, vol. 2964, 2021.
  26. J. Yu, L. Lu, X. Meng, and G. E. Karniadakis, “Gradient-enhanced physics-informed neural networks for forward and inverse pde problems,” Computer Methods in Applied Mechanics and Engineering, vol. 393, p. 114823, 2022.
  27. C. Wu, M. Zhu, Q. Tan, Y. Kartha, and L. Lu, “A comprehensive study of non-adaptive and residual-based adaptive sampling for physics-informed neural networks,” Computer Methods in Applied Mechanics and Engineering, vol. 403, p. 115671, 2023.
  28. L. Yang, X. Meng, and G. E. Karniadakis, “B-pinns: Bayesian physics-informed neural networks for forward and inverse pde problems with noisy data,” Journal of Computational Physics, vol. 425, p. 109913, 2021.
  29. R. Li, J.-X. Wang, E. Lee, and T. Luo, “Physics-informed deep learning for solving phonon boltzmann transport equation with large temperature non-equilibrium,” npj Computational Materials, vol. 8, no. 1, p. 29, 2022.
  30. S. Cai, H. Li, F. Zheng, F. Kong, M. Dao, G. E. Karniadakis, and S. Suresh, “Artificial intelligence velocimetry and microaneurysm-on-a-chip for three-dimensional analysis of blood flow in physiology and disease,” Proceedings of the National Academy of Sciences, vol. 118, no. 13, p. e2100697118, 2021.
  31. N. Wandel, M. Weinmann, and R. Klein, “Teaching the incompressible navier–stokes equations to fast neural surrogate models in three dimensions,” Physics of Fluids, vol. 33, no. 4, p. 047117, 2021.
  32. M. Takamoto, T. Praditia, R. Leiteritz, D. MacKinlay, F. Alesiani, D. Pflüger, and M. Niepert, “PDEBench: An Extensive Benchmark for Scientific Machine Learning,” in 36th Conference on Neural Information Processing Systems (NeurIPS 2022) Track on Datasets and Benchmarks, 2022.
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now