Journal article 90 views 17 downloads
Diffusion models as stochastic quantization in lattice field theory
,
K. Zhou
Journal of High Energy Physics, Volume: 2024, Issue: 5
Swansea University Author:
Gert Aarts
-
PDF | Version of Record
This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0).
Download (1.38MB)
DOI (Published version): 10.1007/jhep05(2024)060
Abstract
In this work, we establish a direct connection between generative diffusion models (DMs) and stochastic quantization (SQ). The DM is realized by approximating the reversal of a stochastic process dictated by the Langevin equation, generating samples from a prior distribution to effectively mimic the...
| Published in: | Journal of High Energy Physics |
|---|---|
| ISSN: | 1029-8479 |
| Published: |
Springer Science and Business Media LLC
2024
|
| Online Access: |
Check full text
|
| URI: | https://cronfa.swan.ac.uk/Record/cronfa66440 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
In this work, we establish a direct connection between generative diffusion models (DMs) and stochastic quantization (SQ). The DM is realized by approximating the reversal of a stochastic process dictated by the Langevin equation, generating samples from a prior distribution to effectively mimic the target distribution. Using numerical simulations, we demonstrate that the DM can serve as a global sampler for generating quantum lattice field configurations in two-dimensional φ4 theory. We demonstrate that DMs can notably reduce autocorrelation times in the Markov chain, especially in the critical region where standard Markov Chain Monte-Carlo (MCMC) algorithms experience critical slowing down. The findings can potentially inspire further advancements in lattice field theory simulations, in particular in cases where it is expensive to generate large ensembles. |
|---|---|
| Keywords: |
Algorithms and Theoretical Developments; Lattice Quantum Field Theory; Non-Perturbative Renormalization; Stochastic Processes |
| College: |
Faculty of Science and Engineering |
| Funders: |
SCOAP3 |
| Issue: |
5 |