BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260924T174609EDT-9936IddVG8@132.216.98.100
DTSTAMP:20260924T214609Z
DESCRIPTION:Title: An implicit tour of regularization\n\nAbstract: Regulari
 zation is a key ingredient in the design of learning algorithms. Classical
 ly it amounts to the definition of a constrained/penalized empirical objec
 tive to be minimized. Optimization aspects are then considered separately.
  In practice\, these distinctions are much more blurred. Indeed\, it is a 
 classical observation that an optimization process can have a self-regular
 izing effect by (implicitly) enforcing some inductive bias.  This observat
 ion has recently become popular in machine learning. On the one hand\, it 
 seems to help understanding learning curves in deep learning. On the other
  hand\, controlling regularization by optimization can improve efficiency 
 in learning. In this talk\, I will provide an overview of classical and re
 cent results on the topic. \n\n \n\n \n\nSeminar MTL Machine Learning and 
 Optimization (MTL MLOpt)\n	Veuillez vous inscrire à la liste d'envoi/Please
  subscribe to the mailing list: https://mtl-mlopt.github.io/\n\nhttps://mt
 l-mlopt.github.io\n
DTSTART:20210210T193000Z
DTEND:20210210T203000Z
SUMMARY:Lorenzo Rosasco (DIBRIS)
URL:https://www.mcgill.ca/mathstat/channels/event/lorenzo-rosasco-dibris-32
 8376
END:VEVENT
END:VCALENDAR