Stochastic path sampler for lattice field theory

A bstract In lattice field theory, target distributions are known only up to normalization, $$ \overset{\sim }{\pi}\left(\phi \right)\propto {e}^{-S\left(\phi \right)} $$ π ~ ϕ ∝ e − S ϕ , while the partition function is generally intractable. Markov chain Monte Carlo simulations often become inefficient near phase transitions or the continuum limit due to critical slowing down. In this work, we propose a novel sampler based on nonequilibrium thermodynamics, called Stochastic Path Sampler (SPS), which can generate configurations for the unnormalized target distribution and without a need for training data. The central idea of the SPS is to establish a trajectory-level balance for learnable forward and backward stochastic dynamics between two equilibrium states — the prior and target distributions. This is achieved by minimizing the path-space variational free energy, equivalently an entropy-production upper bound, defined by the log-ratio of forward and auxiliary backward trajectory measures, to enhance the reversibility of the forward and backward processes. The learned forward process provides independent proposals, which are subsequently corrected by an extended-space Independence Metropolis-Hastings step. In two-dimensional ϕ 4 theory, we demonstrate that our neural sampler can achieve the same sampling quality as HMC but with a much shorter autocorrelation time in the pseudocritical region. This sampler offers a novel stochastic-quantization-inspired route to data-free proposal construction for lattice field theory by leveraging a variational free-energy principle derived from path-space irreversibility.

Authors

Institutions

Publication Details

Journal
Journal of High Energy Physics
Published
2026-11-16
DOI
https://doi.org/10.1007/jhep11(2026)164
Citations
12
Primary Topic
Markov Chains and Monte Carlo Methods
Type
article
Field-Weighted Citation Impact
17.51

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Stochastic path sampler for lattice field theory

Biagio Lucini, Gert Aarts A, Shiyang Chen, Moxian Qian et al.
12 citations
Journal of High Energy Physics
Markov Chains and Monte Carlo Methods
17.51
article

Stochastic path sampler for lattice field theory

Biagio Lucini, Gert Aarts A, Shiyang Chen, Moxian Qian, Kai Zhou
article en
12 citations

Abstract

A bstract In lattice field theory, target distributions are known only up to normalization, $$ \overset{\sim }{\pi}\left(\phi \right)\propto {e}^{-S\left(\phi \right)} $$ π ~ ϕ ∝ e − S ϕ , while the partition function is generally intractable. Markov chain Monte Carlo simulations often become inefficient near phase transitions or the continuum limit due to critical slowing down. In this work, we propose a novel sampler based on nonequilibrium thermodynamics, called Stochastic Path Sampler (SPS), which can generate configurations for the unnormalized target distribution and without a need for training data. The central idea of the SPS is to establish a trajectory-level balance for learnable forward and backward stochastic dynamics between two equilibrium states — the prior and target distributions. This is achieved by minimizing the path-space variational free energy, equivalently an entropy-production upper bound, defined by the log-ratio of forward and auxiliary backward trajectory measures, to enhance the reversibility of the forward and backward processes. The learned forward process provides independent proposals, which are subsequently corrected by an extended-space Independence Metropolis-Hastings step. In two-dimensional ϕ 4 theory, we demonstrate that our neural sampler can achieve the same sampling quality as HMC but with a much shorter autocorrelation time in the pseudocritical region. This sampler offers a novel stochastic-quantization-inspired route to data-free proposal construction for lattice field theory by leveraging a variational free-energy principle derived from path-space irreversibility.

Journal of High Energy PhysicsVol. 2026(11)
Queen Mary University of London (GB), Chinese University of Hong Kong (HK), Johannes Gutenberg University Mainz (DE), Swansea University (GB)
Leverhulme Trust, Swansea University, National Natural Science Foundation of China, China Scholarship Council, Chinese University of Hong Kong, Shenzhen University, Science and Technology Facilities Council, European Regional Development Fund
Openalex Percentile: Top 1%
Markov Chains and Monte Carlo Methods
17.51
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon