BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251122T002601EST-48184VVuOv@132.216.98.100
DTSTAMP:20251122T052601Z
DESCRIPTION:Title: A Conformal-Based Two-Sample Conditional Distribution Te
 st.\n\n\n	Abstract:\n\n\nWe consider the problem of testing the equality of
  the conditional distribution of a response variable given a set of covari
 ates between two populations. Such a testing problem is related to transfe
 r learning and causal inference. We develop a nonparametric procedure by c
 ombining recent advances in conformal prediction with some new ingredients
  such as a novel choice of conformity score and data-driven choices of wei
 ght and score functions. To our knowledge\, this is the first successful a
 ttempt of using conformal prediction for testing statistical hypotheses be
 yond exchangeability. The final test statistic reveals a natural connectio
 n between conformal inference and the classical rank-sum test. Our method 
 is suitable for modern machine learning scenarios where the data has high 
 dimensionality and the sample size is large\, and can be effectively combi
 ned with existing classification algorithms to find good weight and score 
 functions. The performance of the proposed method is demonstrated in synth
 etic and real data examples.\n\n\n	Speaker\n\n\nDr. Jing Lei is a Professor
  in the Department of Statistics and Data Science at Carnegie Mellon Unive
 rsity. His research focuses on conformal prediction\, cross-validation\, g
 eneralization error\, networks\, sparse PCA and optimal transport. He is a
 lso interested in applications such as the analysis of single cell RNA-seq
 uencing and multi-omics data.\n\nɬÀï·¬ Statistics Seminar schedule: https
 ://mcgillstat.github.io/\n\nZoom link\n\n\n	\n		\n			https://mcgill.zoom.us/j/83
 436686293?pwd=b0RmWmlXRXE3OWR6NlNIcWF5d0dJQT09\n\n			Meeting ID: 834 3668 629
 3\n\n			Passcode: 12345\n\n			 \n		\n	\n\n
DTSTART:20221021T193000Z
DTEND:20221021T203000Z
SUMMARY:Jing Lei (Carnegie Mellon University)
URL:/mathstat/channels/event/jing-lei-carnegie-mellon-
 university-342867
END:VEVENT
END:VCALENDAR