Deep multi-step mixed algorithm for high dimensional non-linear PDEs and associated BSDEs

TL;DR

This paper introduces a multistep deep learning algorithm for solving high-dimensional nonlinear BSDEs and PDEs, improving accuracy and reducing complexity compared to existing methods.

Contribution

The paper presents a novel multistep deep learning approach that enhances accuracy and efficiency in solving high-dimensional nonlinear BSDEs and PDEs.

Findings

Increased accuracy over single-step models

Reduced computational complexity

Effective approximation with low regularity conditions

Abstract

We propose a new multistep deep learning-based algorithm for the resolution of moderate to high dimensional nonlinear backward stochastic differential equations (BSDEs) and their corresponding parabolic partial differential equations (PDE). Our algorithm relies on the iterated time discretisation of the BSDE and approximates its solution and gradient using deep neural networks and automatic differentiation at each time step. The approximations are obtained by sequential minimisation of local quadratic loss functions at each time step through stochastic gradient descent. We provide an analysis of approximation error in the case of a network architecture with weight constraints requiring only low regularity conditions on the generator of the BSDE. The algorithm increases accuracy from its single step parent model and has reduced complexity when compared to similar models in the literature.