Sampling in High-Dimensions using Stochastic Interpolants and Forward-Backward Stochastic Differential Equations

TL;DR

This paper introduces a novel diffusion-based sampling method for high-dimensional distributions, leveraging stochastic interpolants and forward-backward stochastic differential equations, with machine learning to solve complex PDEs.

Contribution

It develops a new class of algorithms that transport samples from Gaussian to target distributions using stochastic interpolants and FBSDEs, addressing high-dimensional sampling challenges.

Findings

Effective sampling from complex high-dimensional distributions

Outperforms traditional methods in challenging scenarios

Utilizes machine learning to solve PDEs efficiently

Abstract

We present a class of diffusion-based algorithms to draw samples from high-dimensional probability distributions given their unnormalized densities. Ideally, our methods can transport samples from a Gaussian distribution to a specified target distribution in finite time. Our approach relies on the stochastic interpolants framework to define a time-indexed collection of probability densities that bridge a Gaussian distribution to the target distribution. Subsequently, we derive a diffusion process that obeys the aforementioned probability density at each time instant. Obtaining such a diffusion process involves solving certain Hamilton-Jacobi-Bellman PDEs. We solve these PDEs using the theory of forward-backward stochastic differential equations (FBSDE) together with machine learning-based methods. Through numerical experiments, we demonstrate that our algorithm can effectively draw samples from distributions that conventional methods struggle to handle.