Two-dimensional Parallel Tempering for Constrained Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Delacour, Corentin, Sajeeb, M Mahmudul Hasan, Hespanha, Joao P., Camsari, Kerem Y.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915417988005888
author Delacour, Corentin
Sajeeb, M Mahmudul Hasan
Hespanha, Joao P.
Camsari, Kerem Y.
author_facet Delacour, Corentin
Sajeeb, M Mahmudul Hasan
Hespanha, Joao P.
Camsari, Kerem Y.
contents Sampling Boltzmann probability distributions plays a key role in machine learning and optimization, motivating the design of hardware accelerators such as Ising machines. While the Ising model can in principle encode arbitrary optimization problems, practical implementations are often hindered by soft constraints that either slow down mixing when too strong, or fail to enforce feasibility when too weak. We introduce a two-dimensional extension of the powerful parallel tempering algorithm (PT) that addresses this challenge by adding a second dimension of replicas interpolating the penalty strengths. This scheme ensures constraint satisfaction in the final replicas, analogous to low-energy states at low temperature. The resulting two-dimensional parallel tempering algorithm (2D-PT) improves mixing in heavily constrained replicas and eliminates the need to explicitly tune the penalty strength. In a representative example of graph sparsification with copy constraints, 2D-PT achieves near-ideal mixing, with Kullback-Leibler divergence decaying as O(1/t). When applied to sparsified Wishart instances, 2D-PT yields orders of magnitude speedup over conventional PT with the same number of replicas. The method applies broadly to constrained Ising problems and can be deployed on existing Ising machines.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14781
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Two-dimensional Parallel Tempering for Constrained Optimization
Delacour, Corentin
Sajeeb, M Mahmudul Hasan
Hespanha, Joao P.
Camsari, Kerem Y.
Machine Learning
Statistical Mechanics
Optimization and Control
Sampling Boltzmann probability distributions plays a key role in machine learning and optimization, motivating the design of hardware accelerators such as Ising machines. While the Ising model can in principle encode arbitrary optimization problems, practical implementations are often hindered by soft constraints that either slow down mixing when too strong, or fail to enforce feasibility when too weak. We introduce a two-dimensional extension of the powerful parallel tempering algorithm (PT) that addresses this challenge by adding a second dimension of replicas interpolating the penalty strengths. This scheme ensures constraint satisfaction in the final replicas, analogous to low-energy states at low temperature. The resulting two-dimensional parallel tempering algorithm (2D-PT) improves mixing in heavily constrained replicas and eliminates the need to explicitly tune the penalty strength. In a representative example of graph sparsification with copy constraints, 2D-PT achieves near-ideal mixing, with Kullback-Leibler divergence decaying as O(1/t). When applied to sparsified Wishart instances, 2D-PT yields orders of magnitude speedup over conventional PT with the same number of replicas. The method applies broadly to constrained Ising problems and can be deployed on existing Ising machines.
title Two-dimensional Parallel Tempering for Constrained Optimization
topic Machine Learning
Statistical Mechanics
Optimization and Control
url https://arxiv.org/abs/2506.14781