Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 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

Structure-Preserving Operator Learning: Modeling the Collision Operator of Kinetic Equations (2402.16613v1)

Published 26 Feb 2024 in math.NA and cs.NA

Abstract: This work explores the application of deep operator learning principles to a problem in statistical physics. Specifically, we consider the linear kinetic equation, consisting of a differential advection operator and an integral collision operator, which is a powerful yet expensive mathematical model for interacting particle systems with ample applications, e.g., in radiation transport. We investigate the capabilities of the Deep Operator network (DeepONet) approach to modelling the high dimensional collision operator of the linear kinetic equation. This integral operator has crucial analytical structures that a surrogate model, e.g., a DeepONet, needs to preserve to enable meaningful physical simulation. We propose several DeepONet modifications to encapsulate essential structural properties of this integral operator in a DeepONet model. To be precise, we adapt the architecture of the trunk-net so the DeepONet has the same collision invariants as the theoretical kinetic collision operator, thus preserving conserved quantities, e.g., mass, of the modeled many-particle system. Further, we propose an entropy-inspired data-sampling method tailored to train the modified DeepONet surrogates without requiring an excessive expensive simulation-based data generation.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (39)
  1. A regularized entropy-based moment method for kinetic equations. SIAM Journal on Applied Mathematics, 79(5):1627–1653, 2019.
  2. Babovsky, H. et al. Die Boltzmann-Gleichung: Modellbildung-Numerik-Anwendungen. Springer, 1998.
  3. A neural network closure for the euler-poisson system based on kinetic simulations, 2020.
  4. Two-dimensional time dependent riemann solvers for neutron transport. Journal of Computational Physics, 210(1):386–399, 2005.
  5. Deepm&mnet: Inferring the electroconvection multiphysics fields based on operator approximation by neural networks. Journal of Computational Physics, 436:110296, 2021.
  6. Cercignani, C. The Boltzmann Equation and Its Applications. Springer, New York, NY, 1988.
  7. Chahine, M. T. Foundations of radiation hydrodynamics (dimitri mihalas and barbara weibel mihalas). Siam Review, 29:648–650, 1987.
  8. Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems. IEEE Transactions on Neural Networks, 6(4):911–917, 1995.
  9. Integrating machine learning with physics-based modeling, 2020. URL https://arxiv.org/abs/2006.02619.
  10. Hadorn, P. S. Shift-DeepONet: Extending Deep Operator Networks for Discontinuous Output Functions. ETH Zurich, Seminar for Applied Mathematics, 2022.
  11. Uniformly accurate machine learning-based hydrodynamic models for kinetic equations. Proceedings of the National Academy of Sciences, 116(44):21983–21991, 2019.
  12. Convex duality and entropy-based moment closures: Characterizing degenerate densities. 2008 47th IEEE Conference on Decision and Control, pp. 5092–5097, 2008.
  13. Diffuse radiation in the galaxy. Astrophysical Journal, 93:70–83, January 1941. doi: 10.1086/144246.
  14. Machine learning moment closure models for the radiative transfer equation i: directly learning a gradient based closure. Journal of Computational Physics, pp.  110941, 2022.
  15. Junk, M. Maximum entropy for reduced moment problems. 1999.
  16. Adam: A method for stochastic optimization, 2017.
  17. Kit-rt: An extendable framework for radiative transfer and therapy. ACM Trans. Math. Softw., 49(4), dec 2023. ISSN 0098-3500. doi: 10.1145/3630001. URL https://doi.org/10.1145/3630001.
  18. Error estimates for deeponets: A deep learning framework in infinite dimensions. Transactions of Mathematics and Its Applications, 6(1):tnac001, 2022.
  19. HyperDeepONet: learning operator with complex target function space using the limited resources via hypernetwork. In The Eleventh International Conference on Learning Representations, 2023a. URL https://openreview.net/forum?id=OAw6V3ZAhSd.
  20. opPINN: physics-informed neural network with operator learning to approximate solutions to the Fokker-Planck-Landau equation. J. Comput. Phys., 480:Paper No. 112031, 21, 2023b. ISSN 0021-9991,1090-2716. doi: 10.1016/j.jcp.2023.112031. URL https://doi.org/10.1016/j.jcp.2023.112031.
  21. Computational methods of neutron transport. John Wiley and Sons, Inc, 1984.
  22. Physics-informed neural networks for solving multiscale mode-resolved phonon boltzmann transport equation. Materials Today Physics, 19:100429, 2021a.
  23. Learning invariance preserving moment closure model for boltzmann-bgk equation, 2021b. URL https://arxiv.org/abs/2110.03682.
  24. Operator learning for predicting multiscale bubble growth dynamics. The Journal of Chemical Physics, 154(10):104118, 2021.
  25. Physics-informed neural networks for solving forward and inverse flow problems via the boltzmann-bgk formulation. Journal of Computational Physics, 447:110676, 2021.
  26. Deeponet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators. arXiv preprint arXiv:1910.03193, 2019.
  27. Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators. Nature Machine Intelligence, 3(3):218–229, 2021.
  28. A comprehensive and fair comparison of two neural operators (with practical extensions) based on fair data. Computer Methods in Applied Mechanics and Engineering, 393:114778, 2022. ISSN 0045-7825. doi: https://doi.org/10.1016/j.cma.2022.114778. URL https://www.sciencedirect.com/science/article/pii/S0045782522001207.
  29. DeepM&Mnet for hypersonics: predicting the coupled flow and finite-rate chemistry behind a normal shock using neural-network approximation of operators. J. Comput. Phys., 447:Paper No. 110698, 24, 2021. ISSN 0021-9991. doi: 10.1016/j.jcp.2021.110698. URL https://doi.org/10.1016/j.jcp.2021.110698.
  30. Semiconductor equations. Springer Science & Business Media, 2012.
  31. Neural network representability of fully ionized plasma fluid model closures. Physics of Plasmas, 27(7):072106, 2020.
  32. Machine-learning-based spectral methods for partial differential equations. Scientific Reports, 13(1):1739, 2023.
  33. Physics informed neural networks for simulating radiative transfer. Journal of Quantitative Spectroscopy and Radiative Transfer, 270:107705, 2021.
  34. Structure preserving neural networks: A case study in the entropy closure of the boltzmann equation. In Chaudhuri, K., Jegelka, S., Song, L., Szepesvari, C., Niu, G., and Sabato, S. (eds.), Proceedings of the 39th International Conference on Machine Learning, volume 162 of Proceedings of Machine Learning Research, pp.  19406–19433. PMLR, 17–23 Jul 2022. URL https://proceedings.mlr.press/v162/schotthofer22a.html.
  35. A structure-preserving surrogate model for the closure of the moment system of the Boltzmann equation using convex deep neural networks. 2021. doi: 10.2514/6.2021-2895. URL https://arc.aiaa.org/doi/abs/10.2514/6.2021-2895.
  36. Learning the solution operator of parametric partial differential equations with physics-informed deeponets. Science advances, 7(40):eabi8605, 2021.
  37. Using neural networks to accelerate the solution of the boltzmann equation. Journal of Computational Physics, 443:110521, 2021.
  38. Relaxnet: A structure-preserving neural network to approximate the boltzmann collision operator. Journal of Computational Physics, pp.  112317, 2023.
  39. Predicting continuum breakdown with deep neural networks. Journal of Computational Physics, 489:112278, 2023. ISSN 0021-9991. doi: https://doi.org/10.1016/j.jcp.2023.112278. URL https://www.sciencedirect.com/science/article/pii/S002199912300373X.
Citations (3)
View on Semantic Scholar

Summary

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

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