[main] [papers] [OpenSheaf]mg|pages

On the variational method for Euclidean quantum fields in infinite volume

Nikolay Barashkov and Massimiliano Gubinelli

First issued 9 December 2021 · PDF · HTML

Abstract

We investigate the infinite volume limit of the variational description of Euclidean quantum fields introduced in a previous work. Focussing on two dimensional theories for simplicity, we prove in details how to use the variational approach to obtain tightness of $\varphi^4_2$ without cutoffs and a corresponding large deviation principle for any infinite volume limit. Any infinite volume measure is described via a forward–backwards stochastic differential equation in weak form (wFBSDE). Similar considerations apply to more general $P (\varphi)_2$ theories. We consider also the $\exp (\beta \varphi)_2$ model for $\beta^2 < 8 \pi$ (the so called full $L^1$ regime) and prove uniqueness of the infinite volume limit and a variational characterization of the unique infinite volume measure. The corresponding characterization for $P (\varphi)_2$ theories is lacking due to the difficulty of studying the stability of the wFBSDE against local perturbations.

Versions

  1. Version 2 (arXiv v2) current
    21 November 2022
    arXiv comments: 39 pages, some corrections and remarks on the FBSDE formulation
  2. Version 1 (arXiv v1)
    9 December 2021
    arXiv comments: 34 pages

Cite

Nikolay Barashkov and Massimiliano Gubinelli. On the variational method for Euclidean quantum fields in infinite volume. Papers of Massimiliano Gubinelli, 2021, version 2. https://mgubi.org/papers/paper/variational-infinite-volume/

Licence: arXiv.org perpetual, non-exclusive license to distribute; all other rights reserved to the authors

Pin in the collection (SHA-512 of the inventory it fixes): 2507e42a7ca8a841d7ead096d6da81e02f0d8e2a515ddac7b227a4d17c70169de66ddadf371a2cf9f87a4879ea88b1245d3bd1495ffe5120d056404dd6a05742

Papers of Massimiliano Gubinelli: an OpenSheaf archive (format 0.1, a draft); this page is generated by opensheaf-py from the archive.