Physics ∩ ML

a virtual hub at the interface of theoretical physics and deep learning.

20 May 2020

Building symmetries into generative flow models

Phiala Shanahan, MIT, 12:00 EDT

Abstract: I will discuss recent work to incorporate symmetries, in particular gauge symmetries (local symmetry transformations that form Lie groups), into generative flow models. This work is motivated by the applications of generative models for physics simulation, in particular for lattice field theory.

Based on Equivariant flow-based sampling for lattice gauge theory.

Slides and video of the talk are both available.