BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Elastic methods for curves in two or more dimensions
DTSTART:20230310T151500
DTEND:20230310T170000
DTSTAMP:20260509T225829Z
UID:16095c7d3a81465e5cc1aaf8596a3560e2b047123b2e67df00b5bd60
CATEGORIES:Conferences - Seminars
DESCRIPTION:Sonja Greven\, Humboldt-Universität zu Berlin        
         \nWe provide statistical analysis methods for samples of c
 urves in two or more dimensions\, where only the image but not the parame
 trisation of the curves is of interest. Examples of such data are handwrit
 ten letters\, movement paths or outlines of objects. A parametrisation inv
 ariant analysis can be based on the elastic distance of the curves modulo 
 warping\, but existing methods have limitations in common realistic settin
 gs where curves are irregularly and potentially sparsely observed.\nWe pro
 vide methods and algorithms to approximate the elastic distance for potent
 ially sparsely observed curves\, useful e.g. for classification or cluste
 ring of such curves.\nMoreover\, we propose to use spline curves for model
 ling smooth or polygonal Fréchet means of open or closed curves with resp
 ect to the elastic distance and show identifiability of the spline model m
 odulo warping.\nFinally\, we propose a quotient space regression model for
  elastic regression of such curves on covariates. We test all methods in s
 imulations and apply them to cluster GPS tracks\, classify handwritten spi
 rals of Parkinson's patients and controls\, and to model how the shape of 
 the human hippocampus is related to age\, sex and Alzheimer's disease. \n
 \n 
LOCATION:GA 3 21 https://plan.epfl.ch/?room==GA%203%2021
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR