Speaker
Urs Wenger
(University of Bern)
Description
I will briefly review how machine learning can be used in lattice gauge theory simulations and what approaches are currently available. I will then dicuss one specific application in more detail, namely the machine learning of RG-improved gauge actions using gauge-equivariant convolutional neural networks. In particular, I will present scaling results for a machine-learned fixed-point action in 4d SU(3) gauge theory towards the continuum limit. The results include observables based on the classically perfect gradient-flow scales, which are free of tree-level lattice artifacts to all orders, and quantities related to the static potential and the deconfinement transition.
| Parallel Session (for talks only) | Plenary talk |
|---|
Author
Urs Wenger
(University of Bern)
