BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Uncertainty Quantification in Machine Learning: From Aleatoric to 
 Epistemic - Eyke  Hüllermeier (Ludwig-Maximilians-Universität München)
DTSTART:20250606T130000Z
DTEND:20250606T140000Z
UID:TALK230839@talks.cam.ac.uk
DESCRIPTION:Due to safety requirements in practical applications\, the not
 ion of uncertainty has recently received increasing attention in machine l
 earning research. This talk will address questions regarding the represent
 ation and adequate handling of (predictive) uncertainty in (supervised) ma
 chine learning. A particular focus will be put on the distinction between 
 two important types of uncertainty\, often referred to as aleatoric and ep
 istemic\, and how to quantify these uncertainties in terms of appropriate 
 numerical measures. Roughly speaking\, while aleatoric uncertainty is due 
 to the randomness inherent in the data generating process\, epistemic unce
 rtainty is caused by the learner's ignorance of the true underlying model.
  Some conceptual and theoretical issues of existing methods will identifie
 d\, showing the challenging nature of uncertainty quantification in genera
 l and the disentanglement of aleatoric and epistemic uncertainty in partic
 ular.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR