BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Global and local regression: a signature approach with application
 s
DTSTART:20261110T161500
DTEND:20261110T173000
DTSTAMP:20260924T110831Z
UID:25e2c8d071008b43be9932c8ed4107b3bef7b585d3431fe6406a18b4
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Christian Bayer (WIAS Berlin)\nThe path signature is a p
 owerful tool for solving regression problems on path space\, i.e.\, for co
 mputing conditional expectations $\\mathbb{E}[Y | X]$ when the random vari
 able $X$ is a stochastic process -- or a time-series.\nWe provide new theo
 retical convergence guarantees for two different\, complementary approache
 s to regression using signature methods.\nIn the context of global regress
 ion\, we show that linear functionals of the robust signature are universa
 l in the $L^p$ sense in a wide class of examples.\nIn addition\, we presen
 t a local regression method based on signature semi-metrics\, and show uni
 versality as well as rates of convergence. Based on joint works with Davit
  Gogolashvili\, Luca Pelizzari\, and John Schoenmakers.\n 
LOCATION:CM 1 517 https://plan.epfl.ch/?room==CM%201%20517
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR