Integration Matters for Learning PDEs with Backward SDEs

TL;DR

This paper demonstrates that using Stratonovich-based integration with stochastic Heun methods in BSDEs significantly improves high-dimensional PDE solving, outperforming traditional Euler-Maruyama schemes and rivaling PINNs.

Contribution

It introduces a Stratonovich-based BSDE formulation with stochastic Heun integration to eliminate discretization bias, enhancing the performance of BSDE methods for high-dimensional PDEs.

Findings

Heun-based BSDE outperforms Euler-Maruyama in high dimensions

Stratonovich formulation removes discretization bias

Achieves results competitive with PINNs

Abstract

Backward stochastic differential equation (BSDE)-based deep learning methods provide an alternative to Physics-Informed Neural Networks (PINNs) for solving high-dimensional partial differential equations (PDEs), offering potential algorithmic advantages in settings such as stochastic optimal control, where the PDEs of interest are tied to an underlying dynamical system. However, standard BSDE-based solvers have empirically been shown to underperform relative to PINNs in the literature. In this paper, we identify the root cause of this performance gap as a discretization bias introduced by the standard Euler-Maruyama (EM) integration scheme applied to one-step self-consistency BSDE losses, which shifts the optimization landscape off target. We find that this bias cannot be satisfactorily addressed through finer step-sizes or multi-step self-consistency losses. To properly handle this issue, we propose a Stratonovich-based BSDE formulation, which we implement with stochastic Heun integration. We show that our proposed approach completely eliminates the bias issues faced by EM integration. Furthermore, our empirical results show that our Heun-based BSDE method consistently outperforms EM-based variants and achieves competitive results with PINNs across multiple high-dimensional benchmarks. Our findings highlight the critical role of integration schemes in BSDE-based PDE solvers, an algorithmic detail that has received little attention thus far in the literature.