Diffusion models and stochastic quantisation in lattice field theory

TL;DR

This paper explores the connection between diffusion models and stochastic quantisation in field theory, applying it to generate scalar field configurations on a lattice and discussing potential applications.

Contribution

It introduces a novel approach linking diffusion models with stochastic quantisation for lattice field theory simulations.

Findings

Successfully generated scalar field configurations using the proposed method

Established theoretical connections between diffusion models and stochastic quantisation

Discussed potential applications in field theory simulations

Abstract

Diffusion models are currently the leading generative AI approach used for image generation in e.g. DALL-E and Stable Diffusion. In this talk we relate diffusion models to stochastic quantisation in field theory and employ it to generate configurations for scalar fields on a two-dimensional lattice. We end with some speculations on possible applications.