Saved in:
Bibliographic Details
Main Author: Yallup, David
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.15196
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911276362366976
author Yallup, David
author_facet Yallup, David
contents High-dimensional multimodal sampling problems from lattice field theory (LFT) have become important benchmarks for machine learning assisted sampling methods. We show that GPU-accelerated particle methods, Sequential Monte Carlo (SMC) and nested sampling, provide a strong classical baseline that matches or outperforms state-of-the-art neural samplers in sample quality and wall-clock time on standard scalar field theory benchmarks, while also estimating the partition function. Using only a single data-driven covariance for tuning, these methods achieve competitive performance without problem-specific structure, raising the bar for when learned proposals justify their training cost.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15196
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Particle Monte Carlo methods for Lattice Field Theory
Yallup, David
Machine Learning
High Energy Physics - Lattice
High-dimensional multimodal sampling problems from lattice field theory (LFT) have become important benchmarks for machine learning assisted sampling methods. We show that GPU-accelerated particle methods, Sequential Monte Carlo (SMC) and nested sampling, provide a strong classical baseline that matches or outperforms state-of-the-art neural samplers in sample quality and wall-clock time on standard scalar field theory benchmarks, while also estimating the partition function. Using only a single data-driven covariance for tuning, these methods achieve competitive performance without problem-specific structure, raising the bar for when learned proposals justify their training cost.
title Particle Monte Carlo methods for Lattice Field Theory
topic Machine Learning
High Energy Physics - Lattice
url https://arxiv.org/abs/2511.15196