BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250714T204006EDT-4616AcxFvZ@132.216.98.100
DTSTAMP:20250715T004006Z
DESCRIPTION:Title: Overcoming Statistical Challenges from Multimodal Likeli
 hoods\n\nAbstract: https://people.math.carleton.ca/~davecampbell/Multimoda
 l likelihoods arise when parameters in some parts of the likelihood space 
 at least partially fit the data\, but that there exists a direction in whi
 ch moving parameters degrades the likelihood before eventually rising agai
 n. In differential equation models\, this might be because there are sever
 al ways in which the model dynamics can fit parts of the data. In mixture 
 models\, multimodality might be caused by the models ability to switch mod
 el labels. The main focus of this talk is on methods for overcoming multi-
 modality by incorporating model structure into the algorithm. The talk end
 s with multimodality challenges in inference for modern data types includi
 ng covariate effects from unstructured text and over-parametrization in co
 mputer vision.\n
DTSTART:20200219T203000Z
DTEND:20200219T213000Z
LOCATION:Room 24\, Purvis Hall\, CA\, QC\, Montreal\, H3A 1A2\, 1020 avenue
  des Pins Ouest
SUMMARY:Dave Campbell\, PhD\, Carleton University
URL:/mathstat/channels/event/dave-campbell-phd-carleto
 n-university-320455
END:VEVENT
END:VCALENDAR