Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
98 tokens/sec
GPT-4o
13 tokens/sec
Gemini 2.5 Pro Pro
37 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
3 tokens/sec
DeepSeek R1 via Azure Pro
33 tokens/sec
2000 character limit reached

Exactly conservative physics-informed neural networks and deep operator networks for dynamical systems (2311.14131v1)

Published 23 Nov 2023 in cs.LG, cs.NA, and math.NA

Abstract: We introduce a method for training exactly conservative physics-informed neural networks and physics-informed deep operator networks for dynamical systems. The method employs a projection-based technique that maps a candidate solution learned by the neural network solver for any given dynamical system possessing at least one first integral onto an invariant manifold. We illustrate that exactly conservative physics-informed neural network solvers and physics-informed deep operator networks for dynamical systems vastly outperform their non-conservative counterparts for several real-world problems from the mathematical sciences.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (31)
  1. H. Aref. Point vortex dynamics: a classical mathematics playground. J. Math. Phys., 48(6), 2007.
  2. Invariant physics-informed neural networks for ordinary differential equations. arXiv preprint arXiv:2310.17053, 2023.
  3. A. Bihlo. Improving physics-informed neural networks with meta-learned optimization. arXiv preprint arXiv:2303.07127, 2023.
  4. A. Bihlo and R. O. Popovych. Physics-informed neural networks for the shallow-water equations on the sphere. J. of Comput. Phys., 456:111024, 2022.
  5. A. Bihlo and F. Valiquette. Symmetry-preserving numerical schemes. In Symmetries and Integrability of Difference Equations, pages 261–324. Springer, 2017.
  6. R. Brecht and A. Bihlo. M-ENIAC: A machine learning recreation of the first successful numerical weather forecasts. arXiv preprint arXiv:2304.09070, 2023.
  7. Improving physics-informed DeepONets with hard constraints. arXiv preprint arXiv:2309.07899, 2023.
  8. Language models are few-shot learners. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 1877–1901. Curran Associates, Inc., 2020.
  9. T. Chen and H. Chen. Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems. IEEE Trans. Neural Netw., 6(4):911–917, 1995.
  10. Scientific machine learning through physics–informed neural networks: where we are and what’s next. J. Sci. Comput., 92(3):88, 2022.
  11. A practical method for constructing equivariant multilayer perceptrons for arbitrary matrix groups. In ICML, pages 3318–3328. PMLR, 2021.
  12. H. Goldstein. Classical Mechanics. Addison-Wesley, Reading, 1980.
  13. Geometric numerical integration: structure-preserving algorithms for ordinary differential equations. Springer, Berlin, 2006.
  14. Geometric mechanics and symmetry: from finite to infinite dimensions, volume 12. Oxford University Press, 2009.
  15. Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems. Comput. Methods Appl. Mech. Eng., 365:113028, 2020.
  16. Characterizing possible failure modes in physics-informed neural networks. Advances in Neural Information Processing Systems, 34:26548–26560, 2021.
  17. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, volume 25, pages 1097–1105. Curran Associates, 2012.
  18. Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans. Neural Netw., 9(5):987–1000, 1998.
  19. Double pendulum: An experiment in chaos. Am. J. Phys., 61(11):1038–1044, 1993.
  20. E. N. Lorenz. Deterministic nonperiodic flow. J. Atmos. Sci., 20(2):130–141, 1963.
  21. Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. Nat. Mach. Intell., 3(3):218–229, 2021.
  22. E. H. Müller. Exact conservation laws for neural network integrators of dynamical systems. J. Comput. Phys., 488:112234, 2023.
  23. Y. Nambu. Generalized Hamiltonian dynamics. Phys. Rev. D, 7(8):2405–2412, 1973.
  24. P. Névir and R. Blender. Hamiltonian and Nambu representation of the non-dissipative Lorenz equations. Beitr. Phys. Atmosph., 67(2):133–140, 1994.
  25. M. Raissi. Deep hidden physics models: Deep learning of nonlinear partial differential equations. J. Mach. Learn. Res., 19(1):932–955, 2018.
  26. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys., 378:686–707, 2019.
  27. Numerical Hamiltonian problems, volume 7 of Applied Mathematics and Mathematical Computation. Chapman & Hall, London, 1994.
  28. Mastering the game of Go without human knowledge. Nature, 550(7676):354–359, 2017.
  29. Conservative methods for dynamical systems. SIAM J. Numer. Anal., 55(5):2255–2285, 2017.
  30. S. Wang and P. Perdikaris. Long-time integration of parametric evolution equations with physics-informed DeepONets. J. Comput. Phys., 475:111855, 2023.
  31. Respecting causality is all you need for training physics-informed neural networks. arXiv preprint arXiv:2203.07404, 2022.
Citations (5)

Summary

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

Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.