BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Math Graduate Colloquium: Wasserstein Regression Statistical I
 nference for Distribution Regression with Bures-Wasserstein Metrics
X-WR-TIMEZONE:Pacific Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260820T004618Z
UID:tag:localist.com\,2008:EventInstance_49350126061197
DTSTART:20250416T000000Z
DTEND:20250416T010000Z
DESCRIPTION:Inga Girshfeld\, USC\n\n\nTitle: Wasserstein Regression Statist
 ical Inference for Distribution Regression with Bures-Wasserstein Metrics\
 n\n\nAbstract: Leveraging the structure of the Bures-Wasserstein space\, w
 hich characterizes Gaussian measures through their means and covariance ma
 trices\, we derive a novel estimator for the regression map. The incorpora
 tion of a kernel function in the loss function assigns weights to source d
 istributions based on a Bures-Wasserstein metric\, effectively transformin
 g the problem into a classical kernel regression-like framework. We prove 
 consistency and establish asymptotic normality for the proposed estimator.
GEO:34.022409;-118.291027
LOCATION:Kaprielian Hall (KAP)\, 145
SUMMARY:Math Graduate Colloquium: Wasserstein Regression Statistical Infere
 nce for Distribution Regression with Bures-Wasserstein Metrics
URL;VALUE=URI:https://calendar.usc.edu/event/math-graduate-colloquium-wasse
 rstein-regression-statistical-inference-for-distribution-regression-with-b
 ures-wasserstein-metrics
CATEGORIES:Lecture / Talk / Workshop
END:VEVENT
END:VCALENDAR