Variance reduction in lattice QCD observables via normalizing flows
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908862425071616 |
|---|---|
| author | Abbott, Ryan Boyda, Denis Fu, Yang Hackett, Daniel C. Kanwar, Gurtej Romero-López, Fernando Shanahan, Phiala E. Urban, Julian M. |
| author_facet | Abbott, Ryan Boyda, Denis Fu, Yang Hackett, Daniel C. Kanwar, Gurtej Romero-López, Fernando Shanahan, Phiala E. Urban, Julian M. |
| contents | Normalizing flows can be used to construct unbiased, reduced-variance estimators for lattice field theory observables that are defined by a derivative with respect to action parameters. This work implements the approach for observables involving gluonic operator insertions in the SU(3) Yang-Mills theory and two-flavor Quantum Chromodynamics (QCD) in four space-time dimensions. Variance reduction by factors of $10$-$60$ is achieved in glueball correlation functions and in gluonic matrix elements related to hadron structure, with demonstrated computational advantages. The observed variance reduction is found to be approximately independent of the lattice volume, so that volume transfer can be utilized to minimize training costs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_02984 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Variance reduction in lattice QCD observables via normalizing flows Abbott, Ryan Boyda, Denis Fu, Yang Hackett, Daniel C. Kanwar, Gurtej Romero-López, Fernando Shanahan, Phiala E. Urban, Julian M. High Energy Physics - Lattice Machine Learning Normalizing flows can be used to construct unbiased, reduced-variance estimators for lattice field theory observables that are defined by a derivative with respect to action parameters. This work implements the approach for observables involving gluonic operator insertions in the SU(3) Yang-Mills theory and two-flavor Quantum Chromodynamics (QCD) in four space-time dimensions. Variance reduction by factors of $10$-$60$ is achieved in glueball correlation functions and in gluonic matrix elements related to hadron structure, with demonstrated computational advantages. The observed variance reduction is found to be approximately independent of the lattice volume, so that volume transfer can be utilized to minimize training costs. |
| title | Variance reduction in lattice QCD observables via normalizing flows |
| topic | High Energy Physics - Lattice Machine Learning |
| url | https://arxiv.org/abs/2603.02984 |