Speaker
Description
Recently, the introduction of relevant physical information into neural network architectures has become a widely used and successful strategy for improving the network's performances. In lattice field theories, such information can be identified with gauge symmetries, which are incorporated into the network layers of our recently proposed Lattice Gauge Equivariant Convolutional Neural Networks (L-CNNs) [1]. Previously, we showed how L-CNNs can generalize better to differently sized lattices than traditional neural networks and that they are robust against adversarial gauge transformations.
In this talk, we present our progress on possible applications of L-CNNs as a tool to modify gauge field configurations such as Wilson flow or normalizing flows.
Our methods are based on ordinary differential equations (ODE) parametrized by neural networks (neural ODEs) which allow us to modify link configurations in a gauge equivariant manner.
For simplicity, we focus on simple toy models to test these ideas in practice.
[1] M. Favoni, A. Ipp, D. I. Müller, D. Schuh, Phys.Rev.Lett. 128 (2022) 3, 3 [arXiv:2012.12901]
