BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260717T083854EDT-5727hlNckl@132.216.98.100
DTSTAMP:20260717T123854Z
DESCRIPTION:\n	Abstract:\n\n\nCausal discovery procedures are popular method
 s for discovering causal structure across the physical\, biological\, and 
 social sciences. However\, most procedures for causal discovery only outpu
 t a single estimated causal model or single equivalence class of models. W
 e propose a procedure for quantifying uncertainty in causal discovery. Spe
 cifically\, we consider linear structural equation models with non-Gaussia
 n errors and propose a procedure which returns a confidence sets of causal
  orderings which are not ruled out by the data. We show that asymptoticall
 y\, the true causal ordering will be contained in the returned set with so
 me user specified probability.\n\nJoint work with Sam Wang and Mathias Dar
 ton.\n\n\n	Speaker\n\n\nMladen Kolar is Associate Professor of Econometrics
  and Statistics at the University of Chicago Booth School of Business. Kol
 ar’s research is focused on high-dimensional statistical methods\, probabi
 listic graphical models\, and scalable optimization methods\, driven by th
 e need to uncover interesting and scientifically meaningful structures fro
 m observational data. His research appears in journals such as the Journal
  of Machine Learning Research\, the Annals of Statistics\, the Journal of 
 the Royal Statistical Society\, the Journal of the American Statistical As
 sociation\, Biometrika\, and other outlets. Kolar also regularly presents 
 his research at the top machine learning conferences\, including Advances 
 in Neural Information Processing Systems (NeurIPS) and the International C
 onference of Machine Learning (ICML). Kolar currently serves as associate 
 editor for the Journal of Machine Learning Research\, the Journal of Compu
 tational and Graphical Statistics\, and the New England Journal of Statist
 ics in Data Science.\n\nKolar was awarded a prestigious Facebook Fellowshi
 p in 2010 for his work on machine learning and network models. He spent a 
 summer with Facebook’s ads optimization team working on a large-scale syst
 em for click-through rate prediction. Kolar earned his PhD in Machine Lear
 ning in 2013 from Carnegie Mellon University\, as well as a diploma in Com
 puter Engineering from the University of Zagreb. For his Ph.D. thesis work
  on “Uncovering Structure in High-Dimensions: Networks and Multi-task Lear
 ning Problems\,” Kolar received from 2014 SIGKDD Dissertation Award honora
 ble mention.\n\nOutside of academia\, Kolar enjoys chess\, running\, cycli
 ng\, and hiking.\n\nOn Zoom only\n\nhttps://mcgill.zoom.us/j/83436686293?p
 wd=b0RmWmlXRXE3OWR6NlNIcWF5d0dJQT09\n\nMeeting ID: 834 3668 6293\n\nPassco
 de: 12345\n\n\n	\n		 \n	\n\n
DTSTART:20230324T193000Z
DTEND:20230324T203000Z
SUMMARY:Mladen Kolar
URL:https://www.mcgill.ca/mathstat/channels/event/mladen-kolar-347225
END:VEVENT
END:VCALENDAR