LeapfrogLayers: A Trainable Framework for Effective Topological Sampling

TL;DR

LeapfrogLayers is a novel invertible neural network framework designed to efficiently sample topological features in 2D lattice gauge theories, outperforming traditional methods in autocorrelation times.

Contribution

The paper introduces LeapfrogLayers, a trainable invertible neural network architecture specifically for topological sampling in lattice gauge theories, with improved efficiency over existing methods.

Findings

Reduced autocorrelation time of topological charge compared to HMC

Demonstrated effective transformation of physical quantities

Open source implementation available

Abstract

We introduce LeapfrogLayers, an invertible neural network architecture that can be trained to efficiently sample the topology of a 2D $U(1)$ lattice gauge theory. We show an improvement in the integrated autocorrelation time of the topological charge when compared with traditional HMC, and look at how different quantities transform under our model. Our implementation is open source, and is publicly available on github at https://github.com/saforem2/l2hmc-qcd.