BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:A statistical physics perspective on the theory of machine learnin
 g
DTSTART;VALUE=DATE-TIME:20260415T133000Z
DTEND;VALUE=DATE-TIME:20260415T150000Z
DTSTAMP;VALUE=DATE-TIME:20260506T063144Z
UID:indico-contribution-355@fisindico.uniandes.edu.co
DESCRIPTION:Speakers: Bruno Loureiro (ENS Paris)\nThe past decade has witn
 essed a surge in the development and adoption of machine learning algorith
 ms to solve day-a-day computational tasks. Yet\, a solid theoretical under
 standing of even the most basic tools used in practice is still lacking\, 
 as traditional statistical learning methods are unfit to deal with the mod
 ern regime in which the number of model parameters are of the same order a
 s the quantity of data – a problem known as the curse of dimensionality.
  Curiously\, this is precisely the regime studied by Physicists since the 
 mid 19th century in the context of interacting many-particle systems. This
  connection\, which was first established in the seminal work of Elisabeth
  Gardner and Bernard Derrida in the 80s\, is the basis of a long and fruit
 ful marriage between these two fields.\n\nThe goal of this mini-course is 
 to provide an in-depth overview of these connections and a good vision of 
 the different tools available in the statistical physics toolbox\, as well
  as their scope and limitations. \n\nSyllabus:\n\n- Historical overview of
  the connections between Statistical Physics and Computer Science.\n- Mean
 -Field Models 101: Curie-Weiss Model\n- Statistical-to-Computational Gaps\
 n- The double-descent phenomena and benign overfitting (time permitting)\n
 \nBibliography:\n\nThe mini-course will the based on the following lecture
 s notes: https://brloureiro.github.io/assets/pdf/NotesPrinceton_BL.pdf\n\n
 https://fisindico.uniandes.edu.co/event/23/contributions/355/
LOCATION:Universidad Nacional Ed. 564
URL:https://fisindico.uniandes.edu.co/event/23/contributions/355/
END:VEVENT
END:VCALENDAR