A learning scheme by sparse grids and Picard approximations for semilinear parabolic PDEs

Abstract

Relying on the classical connection between Backward Stochastic Differential Equations (BSDEs) and non-linear parabolic partial differential equations (PDEs), we propose a new probabilistic learning scheme for solving high-dimensional semi-linear parabolic PDEs. This scheme is inspired by the approach coming from machine learning and developed using deep neural networks in Han and al. [32]. Our algorithm is based on a Picard iteration scheme in which a sequence of linear-quadratic optimisation problem is solved by means of stochastic gradient descent (SGD) algorithm. In the framework of a linear specification of the approximation space, we manage to prove a convergence result for our scheme, under some smallness condition. In practice, in order to be able to treat high-dimensional examples, we employ sparse grid approximation spaces. In the case of periodic coefficients and using pre-wavelet basis functions, we obtain an upper bound on the global complexity of our method. It shows in particular that the curse of dimensionality is tamed in the sense that in order to achieve a root mean squared error of order $\epsilon$, for a prescribed precision $\epsilon$, the complexity of the Picard algorithm grows polynomially in ${\epsilon }^{-1}$ up to some logarithmic factor $|log(\epsilon )|$ which grows linearly with respect to the PDE dimension. Various numerical results are presented to validate the performance of our method and to compare them with some recent machine learning schemes proposed in Han and al. [20] and Huré and al. [37].

References in corpus (15)

• Solving high-dimensional partial differential equations using deep learning
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• A regression-based Monte Carlo method to solve backward stochastic differential equations
• Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations
• Algorithms for Solving High Dimensional PDEs: From Nonlinear Monte Carlo to Machine Learning
• Convergence of the Deep BSDE Method for Coupled FBSDEs
• An overview on deep learning-based approximation methods for partial differential equations
• Simulation of BSDEs by Wiener chaos expansion
• Multilevel Picard iterations for solving smooth semilinear parabolic heat equations
• SelectNet: Self-paced Learning for High-dimensional Partial Differential Equations
• Numerical simulation of quadratic BSDEs
• On some Non Asymptotic Bounds for the Euler Scheme
• Stratified regression Monte-Carlo scheme for semilinear PDEs and BSDEs with large scale parallelization on GPUs
• Approximation of backward stochastic differential equations using Malliavin weights and least-squares regression
• Efficient cosmological parameter sampling using sparse grids