BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Minimum Description Length - Carl E. Rasmussen\; Niki Kilbertus
DTSTART:20180125T133000Z
DTEND:20180125T150000Z
UID:TALK94234@talks.cam.ac.uk
CONTACT:Alessandro Davide Ialongo
DESCRIPTION:*Abstract*\n\n"The Minimum Description Length (MDL) principle 
 is a method for inductive inference that provides a generic solution to th
 e model selection problem\, and\, more generally to the overfitting proble
 m.” (Peter Grünwald)\nIn this talk we will provide a very basic introdu
 ction to the philosophy and general idea behind the MDL principle that vie
 ws learning as data compression. While we focus mostly on the high level g
 oals of MDL and how it compares to\, e.g.\, Bayesian inference or the info
 rmation bottleneck\, we will gently introduce a “crude two-part” versi
 on of MDL in some detail. Finally we briefly outline a “refined” versi
 on of MDL and discuss its pros and cons.\n\n*Recommended Reading*\n\nSince
  we start from scratch\, no reading is required.\nFor a fruitful discussio
 n about MDL vs. Bayesian Inference vs. Frequentist at the end of the talk 
 you can browse (the beginnings of) section 17.1 and 17.2 here:\n\nhttps://
 homepages.cwi.nl/~pdg/book/chapter17.pdf
LOCATION:Engineering Department\, CBL Seminar Room 4-38
END:VEVENT
END:VCALENDAR