KPZ equation from a class of nonlinear SPDEs in infinite volume
Kevin Yang
TL;DR
The paper proves that a broad class of nonlinear Ginzburg–Landau SPDEs in infinite volume, under weak-nonlinearity scaling and non-equilibrium initial data, converge to the KPZ equation on the full line. Its core method is a space-discretization leading to infinite-dimensional SDEs, together with a carefully constructed stochastic heat kernel and a stochastic Cole–Hopf transform, which together tame infinite-volume singularities. The authors establish universality results for both continuous and wedge-type initial data, including a wedge limit that yields Tracy–Widom fluctuations, and extend prior torus results to the real line while handling non-equilibrium data beyond Brownian-type initial data. This approach provides a robust bridge from nonlinear SPDEs to KPZ in infinite volume and connects to random-matrix theory through the wedge/slope universality, with potential implications for non-equilibrium statistical mechanics and integrable probability.
Abstract
We study a general class of nonlinear Ginzburg-Landau SPDEs in infinite volume under weak nonlinearity scaling and with non-equilibrium initial data. We derive the KPZ equation as a continuum limit of these equations. This makes rigorous the original derivation of the KPZ equation from physics in the full-space setting, which was a problem posed by Hairer-Quastel '18. Our analysis is based on a stochastic heat kernel for a linearization of said SPDEs.
