From quadratic harnesses, through Askey-Wilson processes and ASEPs, to identification of the stationary measure of the open KPZ equation on the interval.

Date/heure
24 mars 2022
09:15 - 10:15

Oratrice ou orateur
Jacek Wesolowski (Warsaw University of Technology)

Catégorie d'évènement
Groupe de travail Probabilités et Statistique


Résumé

Quadratic harnesses (QH) are Markov processes with linear
conditional expectations and quadratic conditional variances given the
natural past-future filtration. They are governed by 5 numerical
constants hidden in coefficients of conditional variances. A large
family of QH processes can be identified through Askey-Wilson (AW)
processes, which are Markov processes with transition and marginal laws
defined in terms of orthogonality measures of the celebrated system of
the Askey-Wilson polynomials. We proved in 2017 (joint paper with W.
Bryc) that the generating function for the stationary distribution of
the ASEP (asymmetric simple exclusion process) with open boundaries can
be represented through moments of QH (and AW) processes. I.Corwin and
A.Knizel (2021) used this representation for ASEPs of growing size with
a suitable asymptotic regime to find the Laplace transform of the
stationary measure of the open Kardar-Parisi-Zhang (KPZ_) equation on
the interval. Recently (joint paper with W. Bryc, A. Kuznetsov, Y. Wang)
we « inverted » this Laplace transform and thus identified directly the
solution of the open KPZ in terms of a Doob h-transform of the Brownian
motion killed at an exponential rate.