Adjoint Schrödinger Bridge Sampler

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Guan-Horng, Choi, Jaemoo, Chen, Yongxin, Miller, Benjamin Kurt, Chen, Ricky T. Q.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909922034188288
author Liu, Guan-Horng
Choi, Jaemoo
Chen, Yongxin
Miller, Benjamin Kurt
Chen, Ricky T. Q.
author_facet Liu, Guan-Horng
Choi, Jaemoo
Chen, Yongxin
Miller, Benjamin Kurt
Chen, Ricky T. Q.
contents Computational methods for learning to sample from the Boltzmann distribution -- where the target distribution is known only up to an unnormalized energy function -- have advanced significantly recently. Due to the lack of explicit target samples, however, prior diffusion-based methods, known as diffusion samplers, often require importance-weighted estimation or complicated learning processes. Both trade off scalability with extensive evaluations of the energy and model, thereby limiting their practical usage. In this work, we propose Adjoint Schrödinger Bridge Sampler (ASBS), a new diffusion sampler that employs simple and scalable matching-based objectives yet without the need to estimate target samples during training. ASBS is grounded on a mathematical model -- the Schrödinger Bridge -- which enhances sampling efficiency via kinetic-optimal transportation. Through a new lens of stochastic optimal control theory, we demonstrate how SB-based diffusion samplers can be learned at scale via Adjoint Matching and prove convergence to the global solution. Notably, ASBS generalizes the recent Adjoint Sampling (Havens et al., 2025) to arbitrary source distributions by relaxing the so-called memoryless condition that largely restricts the design space. Through extensive experiments, we demonstrate the effectiveness of ASBS on sampling from classical energy functions, amortized conformer generation, and molecular Boltzmann distributions. Code available at https://github.com/facebookresearch/adjoint_samplers
format Preprint
id arxiv_https___arxiv_org_abs_2506_22565
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adjoint Schrödinger Bridge Sampler
Liu, Guan-Horng
Choi, Jaemoo
Chen, Yongxin
Miller, Benjamin Kurt
Chen, Ricky T. Q.
Machine Learning
Optimization and Control
Computational methods for learning to sample from the Boltzmann distribution -- where the target distribution is known only up to an unnormalized energy function -- have advanced significantly recently. Due to the lack of explicit target samples, however, prior diffusion-based methods, known as diffusion samplers, often require importance-weighted estimation or complicated learning processes. Both trade off scalability with extensive evaluations of the energy and model, thereby limiting their practical usage. In this work, we propose Adjoint Schrödinger Bridge Sampler (ASBS), a new diffusion sampler that employs simple and scalable matching-based objectives yet without the need to estimate target samples during training. ASBS is grounded on a mathematical model -- the Schrödinger Bridge -- which enhances sampling efficiency via kinetic-optimal transportation. Through a new lens of stochastic optimal control theory, we demonstrate how SB-based diffusion samplers can be learned at scale via Adjoint Matching and prove convergence to the global solution. Notably, ASBS generalizes the recent Adjoint Sampling (Havens et al., 2025) to arbitrary source distributions by relaxing the so-called memoryless condition that largely restricts the design space. Through extensive experiments, we demonstrate the effectiveness of ASBS on sampling from classical energy functions, amortized conformer generation, and molecular Boltzmann distributions. Code available at https://github.com/facebookresearch/adjoint_samplers
title Adjoint Schrödinger Bridge Sampler
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2506.22565