Numerical Simulations for Full History Recursive Multilevel Picard Approximations for Systems of High-Dimensional Partial Differential Equations

Numerical Simulations for Full History Recursive Multilevel Picard Approximations for Systems of High-Dimensional Partial Differential Equations

Year:    2020

Author:    Sebastian Becker, Ramon Braunwarth, Martin Hutzenthaler, Arnulf Jentzen, Philippe von Wurstemberger

Communications in Computational Physics, Vol. 28 (2020), Iss. 5 : pp. 2109–2138

Abstract

One of the most challenging issues in applied mathematics is to develop and analyze algorithms which are able to approximately compute solutions of high-dimensional nonlinear partial differential equations (PDEs). In particular, it is very hard to develop approximation algorithms which do not suffer under the curse of dimensionality in the sense that the number of computational operations needed by the algorithm to compute an approximation of accuracy $ε$>0 grows at most polynomially in both the reciprocal 1/$ε$ of the required accuracy and the dimension $d∈\mathbb{N}$ of the PDE. Recently, a new approximation method, the so-called full history recursive multilevel Picard (MLP) approximation method, has been introduced and, until today, this approximation scheme is the only approximation method in the scientific literature which has been proven to overcome the curse of dimensionality in the numerical approximation of semilinear PDEs with general time horizons. It is a key contribution of this article to extend the MLP approximation method to systems of semilinear PDEs and to numerically test it on several example PDEs. More specifically, we apply the proposed MLP approximation method in the case of Allen-Cahn PDEs, Sine-Gordon-type PDEs, systems of coupled semilinear heat PDEs, and semilinear Black-Scholes PDEs in up to 1000 dimensions. We also compare the performance of the proposed MLP approximation algorithm with a deep learning based approximation method from the scientific literature.

You do not have full access to this article.

Already a Subscriber? Sign in as an individual or via your institution

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/cicp.OA-2020-0130

Communications in Computational Physics, Vol. 28 (2020), Iss. 5 : pp. 2109–2138

Published online:    2020-01

AMS Subject Headings:    Global Science Press

Copyright:    COPYRIGHT: © Global Science Press

Pages:    30

Keywords:    Curse of dimensionality partial differential equation PDE high-dimensional semilinear multilevel Picard.

Author Details

Sebastian Becker

Ramon Braunwarth

Martin Hutzenthaler

Arnulf Jentzen

Philippe von Wurstemberger

  1. Deep learning approximations for non-local nonlinear PDEs with Neumann boundary conditions

    Boussange, Victor | Becker, Sebastian | Jentzen, Arnulf | Kuckuck, Benno | Pellissier, Loïc

    Partial Differential Equations and Applications, Vol. 4 (2023), Iss. 6

    https://doi.org/10.1007/s42985-023-00244-0 [Citations: 2]
  2. Algorithms for solving high dimensional PDEs: from nonlinear Monte Carlo to machine learning

    E, Weinan | Han, Jiequn | Jentzen, Arnulf

    Nonlinearity, Vol. 35 (2022), Iss. 1 P.278

    https://doi.org/10.1088/1361-6544/ac337f [Citations: 53]
  3. Numerical solution of the modified and non-Newtonian Burgers equations by stochastic coded trees

    Nguwi, Jiang Yu | Privault, Nicolas

    Japan Journal of Industrial and Applied Mathematics, Vol. 40 (2023), Iss. 3 P.1745

    https://doi.org/10.1007/s13160-023-00611-9 [Citations: 0]
  4. A deep learning approach to the probabilistic numerical solution of path-dependent partial differential equations

    Nguwi, Jiang Yu | Privault, Nicolas

    Partial Differential Equations and Applications, Vol. 4 (2023), Iss. 4

    https://doi.org/10.1007/s42985-023-00255-x [Citations: 0]
  5. A fully nonlinear Feynman–Kac formula with derivatives of arbitrary orders

    Nguwi, Jiang Yu | Penent, Guillaume | Privault, Nicolas

    Journal of Evolution Equations, Vol. 23 (2023), Iss. 1

    https://doi.org/10.1007/s00028-023-00873-3 [Citations: 4]
  6. Overcoming the curse of dimensionality in the numerical approximation of backward stochastic differential equations

    Hutzenthaler, Martin | Jentzen, Arnulf | Kruse, Thomas | Anh Nguyen, Tuan

    Journal of Numerical Mathematics, Vol. 0 (2022), Iss. 0

    https://doi.org/10.1515/jnma-2021-0111 [Citations: 0]
  7. Gradient boosting-based numerical methods for high-dimensional backward stochastic differential equations

    Teng, Long

    Applied Mathematics and Computation, Vol. 426 (2022), Iss. P.127119

    https://doi.org/10.1016/j.amc.2022.127119 [Citations: 3]
  8. A deep branching solver for fully nonlinear partial differential equations

    Nguwi, Jiang Yu | Penent, Guillaume | Privault, Nicolas

    Journal of Computational Physics, Vol. 499 (2024), Iss. P.112712

    https://doi.org/10.1016/j.jcp.2023.112712 [Citations: 2]
  9. Solve High-Dimensional Reflected Partial Differential Equations by Neural Network Method

    Shi, Xiaowen | Zhang, Xiangyu | Tang, Renwu | Yang, Juan

    Mathematical and Computational Applications, Vol. 28 (2023), Iss. 4 P.79

    https://doi.org/10.3390/mca28040079 [Citations: 0]
  10. Numerical solution for high-dimensional partial differential equations based on deep learning with residual learning and data-driven learning

    Wang, Zheng | Weng, Futian | Liu, Jialin | Cao, Kai | Hou, Muzhou | Wang, Juan

    International Journal of Machine Learning and Cybernetics, Vol. 12 (2021), Iss. 6 P.1839

    https://doi.org/10.1007/s13042-021-01277-w [Citations: 7]
  11. Deep Splitting Method for Parabolic PDEs

    Beck, Christian | Becker, Sebastian | Cheridito, Patrick | Jentzen, Arnulf | Neufeld, Ariel

    SIAM Journal on Scientific Computing, Vol. 43 (2021), Iss. 5 P.A3135

    https://doi.org/10.1137/19M1297919 [Citations: 44]
  12. Overcoming the Curse of Dimensionality in the Numerical Approximation of Parabolic Partial Differential Equations with Gradient-Dependent Nonlinearities

    Hutzenthaler, Martin | Jentzen, Arnulf | Kruse, Thomas

    Foundations of Computational Mathematics, Vol. 22 (2022), Iss. 4 P.905

    https://doi.org/10.1007/s10208-021-09514-y [Citations: 10]
  13. Overcoming the curse of dimensionality in the numerical approximation of Allen–Cahn partial differential equations via truncated full-history recursive multilevel Picard approximations

    Beck, Christian | Hornung, Fabian | Hutzenthaler, Martin | Jentzen, Arnulf | Kruse, Thomas

    Journal of Numerical Mathematics, Vol. 28 (2020), Iss. 4 P.197

    https://doi.org/10.1515/jnma-2019-0074 [Citations: 19]
  14. Solving Non-linear Kolmogorov Equations in Large Dimensions by Using Deep Learning: A Numerical Comparison of Discretization Schemes

    Marino, Raffaele | Macris, Nicolas

    Journal of Scientific Computing, Vol. 94 (2023), Iss. 1

    https://doi.org/10.1007/s10915-022-02044-x [Citations: 4]