Abstract:In this work, we propose a novel backward differential deep learning-based algorithm for solving high-dimensional nonlinear backward stochastic differential equations (BSDEs), where the deep neural network (DNN) models are trained not only on the inputs and labels but also the differentials of the corresponding labels. This is motivated by the fact that differential deep learning can provide an efficient approximation of the labels and their derivatives with respect to inputs. The BSDEs are reformulated as differential deep learning problems by using Malliavin calculus. The Malliavin derivatives of solution to a BSDE satisfy themselves another BSDE, resulting thus in a system of BSDEs. Such formulation requires the estimation of the solution, its gradient, and the Hessian matrix, represented by the triple of processes $\left(Y, Z, \Gamma\right).$ All the integrals within this system are discretized by using the Euler-Maruyama method. Subsequently, DNNs are employed to approximate the triple of these unknown processes. The DNN parameters are backwardly optimized at each time step by minimizing a differential learning type loss function, which is defined as a weighted sum of the dynamics of the discretized BSDE system, with the first term providing the dynamics of the process $Y$ and the other the process $Z$. An error analysis is carried out to show the convergence of the proposed algorithm. Various numerical experiments up to $50$ dimensions are provided to demonstrate the high efficiency. Both theoretically and numerically, it is demonstrated that our proposed scheme is more efficient compared to other contemporary deep learning-based methodologies, especially in the computation of the process $\Gamma$.

RNN-BSDE method for high-dimensional fractional backward stochastic differential equations with Wick-Itô integrals

Solutions to Bsdes Driven by Both Standard and Fractional Brownian Motions

A deep learning-based Monte Carlo simulation scheme for stochastic differential equations driven by fractional Brownian motion

Deep learning numerical methods for high-dimensional fully nonlinear PIDEs and coupled FBSDEs with jumps

Numerical Method for Multi-Dimensional Coupled Forward-Backward Stochastic Differential Equations Based on Fractional Fourier Fast Transform

FBSDE based Neural Network Algorithms for High-Dimensional Quasilinear Parabolic PDEs

A backward differential deep learning-based algorithm for solving high-dimensional nonlinear backward stochastic differential equations

The One Step Malliavin scheme: new discretization of BSDEs implemented with deep learning regressions

A Deep Learning Method for Solving Multi-Dimensional Coupled Forward–backward Doubly SDEs

Solutions to BSDEs Driven by Both Fractional Brownian Motions and the Underlying Standard Brownian Motions

Solving High-Dimensional Forward-Backward Doubly SDEs and Their Related SPDEs Through Deep Learning

Numerical solution of nonlinear stochastic differential equations with fractional Brownian motion using fractional-order Genocchi deep neural networks

A forward differential deep learning-based algorithm for solving high-dimensional nonlinear backward stochastic differential equations

Deep Learning-Based Numerical Methods for High-Dimensional Parabolic Partial Differential Equations and Backward Stochastic Differential Equations

Mean-field Backward Stochastic Differential Equations Driven by Fractional Brownian Motion

Deep FBSDE Neural Networks for Solving Incompressible Navier-Stokes Equation and Cahn-Hilliard Equation

Semilinear Backward Doubly Stochastic Differential Equations and SPDEs Driven by Fractional Brownian Motion with Hurst Parameter in (0,1/2)

Solution to Forward-Backward Stochastic Differential Equations with Random Coefficients and Application to Deterministic Optimal Control

A Convolution Method for Numerical Solution of Backward Stochastic Differential Equations Based on the Fractional FFT

Numerical Algorithms for Backward Stochastic Differential Equations with 1-D Brownian Motion: Convergence and Simulations

FBSJNN: A Theoretically Interpretable and Efficiently Deep Learning method for Solving Partial Integro-Differential Equations