BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20250918T001425EDT-37247WhTDN@132.216.98.100 DTSTAMP:20250918T041425Z DESCRIPTION:Title: Random matrices and the Gaussian multiplicative chaos on the line\n \n Abstract: The Gaussian multiplicative chaos is a relatively n ew universal object in probability that has many interesting geometric pro perties.\n\nThe characteristic polynomial of many classes of random matric es is\, in many cases conjecturally\, one class of finite approximation to these random measures. Great progress has been made on showing the random matrices from specific ``circular ensembles’’ converge to the GMC. Likewi se\, some progress has been made for unitarily—invariant random matrices. We show some new partial progress in showing the 'Gaussian beta—ensemble' has a GMC limit. This we do by using the representation of its characteris tic polynomial as an entry in a product of independent random two--by--two matrices. For a point z in the complex plane\, at which the transfer matr ix is to be evaluated\, this product of transfer matrices splits into thre e independent factors\, each of which can be understood as a different dyn amical system in the complex plane. Using this\, we show that the characte ristic polynomial is always represented as product of at most three terms\ , two of which are exponentials of Gaussian fields and one of which is the stochastic Airy function\, up to vanishing multiplicative errors. Suffici ently far into the complex plane\, it can be represented using only one. A t a point z at the spectral edge\, there are two. At a point z in the bulk of the spectrum\, all three are necessary to describe the characteristic polynomial. Joint work with Gaultier Lambert.\n DTSTART:20191009T190000Z DTEND:20191009T200000Z LOCATION:Room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue Sherbrooke Ouest SUMMARY:Elliot Paquette (Ohio State) URL:/mathstat/channels/event/elliot-paquette-ohio-stat e-301320 END:VEVENT END:VCALENDAR