# Gauge\-equivariant neural networks as preconditioners in lattice QCD

**INSPIRE:** [2631052](https://inspirehep.net/literature/2631052)
**arXiv:** [2302\.05419](https://arxiv.org/abs/2302.05419)
**DOI:** [10\.1103/PhysRevD\.108\.034503](https://doi.org/10.1103/PhysRevD.108.034503)

**Authors:** Lehner, Christoph, Wettig, Tilo

**Submitted:** 10 February 2023

**Subjects:**
- hep\-lat
- cs\.LG
- cs\.NA
- math\.NA
- Lattice
- Computing
- Math and Math Physics

**Journal reference:** Phys\.Rev\.D 108 \(3\) 034503 \(2023\)

**Citations:** 25

## Abstract

We demonstrate that a state\-of\-the\-art multigrid preconditioner can be learned efficiently by gauge\-equivariant neural networks\. We show that the models require minimal retraining on different gauge configurations of the same gauge ensemble and to a large extent remain efficient under modest modifications of ensemble parameters\. We also demonstrate that important paradigms such as communication avoidance are straightforward to implement in this framework\.