Testing machine-learned distributions against Monte Carlo data for the QCD chiral phase transition
Abstract
We demonstrate that conditional Masked Autoregressive Flows constitute a flexible interpolation tool for lattice QCD observables, conditioned on bare lattice parameters. As a benchmark, we use the chiral phase structure of QCD with five degenerate light quark flavours, which on coarse lattices exhibits a region of first-order chiral transitions terminating in a critical quark mass. The method successfully reproduces standard reweighting in the gauge coupling, and naturally extends to interpolation in quark mass and spatial volume, for which reweighting is computationally prohibitive or inapplicable, respectively. Once trained, the model generates samples across the full parameter space in minutes, which can be used to obtain consistent first estimates of the critical quark mass without simulating all intermediate parameter values. This offers a concrete reduction in the number of lattice ensembles required. Precision on the critical mass from learned distributions is so far prohibited by the mode-covering effect inherent to maximum-likelihood-based training, which introduces a systematic bias near first-order transitions. At the current stage, the method is well-suited for a range of practical applications: localising phase boundaries, identifying the universal scaling axes at a critical point, and accelerating informed determinations of parameter values ahead of high-precision Monte Carlo campaigns.
Versions
Related
- Estimating redshift distributions of Fermi blazars with multiple machine-learning models
- An atom-based machine-learned dipole-moment model and application to conjugated systems
- Machine-learned spectral PDE solvers for fluids and plasmas
- Ab initio Modeling of MoS2/Oxide Device Interfaces with Machine Learned Electronic Structures
- Capturing the Effect of Exact Exchange in Thio-Michael Additions with a Machine-Learned Functional at Meta-GGA Cost
BibTeX
@misc{kaiser2026testing,
title = {{Testing machine-learned distributions against Monte Carlo data for the QCD chiral phase transition}},
author = {Reinhold Kaiser and Frithjof Karsch and Jan Philipp Klinger and Owe Philipsen and Christian Schmidt and Simran Singh},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2605.07262},
eprint = {2605.07262},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.07262},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Testing machine-learned distributions against Monte Carlo data for the QCD chiral phase transition
AU - Reinhold Kaiser
AU - Frithjof Karsch
AU - Jan Philipp Klinger
AU - Owe Philipsen
AU - Christian Schmidt
AU - Simran Singh
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2605.07262
UR - https://arxiv.org/abs/2605.07262
AB - We demonstrate that conditional Masked Autoregressive Flows constitute a flexible interpolation tool for lattice QCD observables, conditioned on bare lattice parameters. As a benchmark, we use the chiral phase structure of QCD with five degenerate light quark flavours, which on coarse lattices exhibits a region of first-order chiral transitions terminating in a critical quark mass. The method successfully reproduces standard reweighting in the gauge coupling, and naturally extends to interpolation in quark mass and spatial volume, for which reweighting is computationally prohibitive or inapplicable, respectively. Once trained, the model generates samples across the full parameter space in minutes, which can be used to obtain consistent first estimates of the critical quark mass without simulating all intermediate parameter values. This offers a concrete reduction in the number of lattice ensembles required. Precision on the critical mass from learned distributions is so far prohibited by the mode-covering effect inherent to maximum-likelihood-based training, which introduces a systematic bias near first-order transitions. At the current stage, the method is well-suited for a range of practical applications: localising phase boundaries, identifying the universal scaling axes at a critical point, and accelerating informed determinations of parameter values ahead of high-precision Monte Carlo campaigns.
ER -