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.
- H. Aref. Point vortex dynamics: a classical mathematics playground. J. Math. Phys., 48(6), 2007.
- Invariant physics-informed neural networks for ordinary differential equations. arXiv preprint arXiv:2310.17053, 2023.
- A. Bihlo. Improving physics-informed neural networks with meta-learned optimization. arXiv preprint arXiv:2303.07127, 2023.
- 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.
- A. Bihlo and F. Valiquette. Symmetry-preserving numerical schemes. In Symmetries and Integrability of Difference Equations, pages 261–324. Springer, 2017.
- R. Brecht and A. Bihlo. M-ENIAC: A machine learning recreation of the first successful numerical weather forecasts. arXiv preprint arXiv:2304.09070, 2023.
- Improving physics-informed DeepONets with hard constraints. arXiv preprint arXiv:2309.07899, 2023.
- 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.
- 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.
- Scientific machine learning through physics–informed neural networks: where we are and what’s next. J. Sci. Comput., 92(3):88, 2022.
- A practical method for constructing equivariant multilayer perceptrons for arbitrary matrix groups. In ICML, pages 3318–3328. PMLR, 2021.
- H. Goldstein. Classical Mechanics. Addison-Wesley, Reading, 1980.
- Geometric numerical integration: structure-preserving algorithms for ordinary differential equations. Springer, Berlin, 2006.
- Geometric mechanics and symmetry: from finite to infinite dimensions, volume 12. Oxford University Press, 2009.
- 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.
- Characterizing possible failure modes in physics-informed neural networks. Advances in Neural Information Processing Systems, 34:26548–26560, 2021.
- Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, volume 25, pages 1097–1105. Curran Associates, 2012.
- Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans. Neural Netw., 9(5):987–1000, 1998.
- Double pendulum: An experiment in chaos. Am. J. Phys., 61(11):1038–1044, 1993.
- E. N. Lorenz. Deterministic nonperiodic flow. J. Atmos. Sci., 20(2):130–141, 1963.
- Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. Nat. Mach. Intell., 3(3):218–229, 2021.
- E. H. Müller. Exact conservation laws for neural network integrators of dynamical systems. J. Comput. Phys., 488:112234, 2023.
- Y. Nambu. Generalized Hamiltonian dynamics. Phys. Rev. D, 7(8):2405–2412, 1973.
- P. Névir and R. Blender. Hamiltonian and Nambu representation of the non-dissipative Lorenz equations. Beitr. Phys. Atmosph., 67(2):133–140, 1994.
- M. Raissi. Deep hidden physics models: Deep learning of nonlinear partial differential equations. J. Mach. Learn. Res., 19(1):932–955, 2018.
- 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.
- Numerical Hamiltonian problems, volume 7 of Applied Mathematics and Mathematical Computation. Chapman & Hall, London, 1994.
- Mastering the game of Go without human knowledge. Nature, 550(7676):354–359, 2017.
- Conservative methods for dynamical systems. SIAM J. Numer. Anal., 55(5):2255–2285, 2017.
- S. Wang and P. Perdikaris. Long-time integration of parametric evolution equations with physics-informed DeepONets. J. Comput. Phys., 475:111855, 2023.
- Respecting causality is all you need for training physics-informed neural networks. arXiv preprint arXiv:2203.07404, 2022.