Upcoming Talks

Ist logo

Statistical learning theory from a geometric viewpoint

Date: Thursday, April 5, 2018 09:00 - 10:00
Speaker: Shahar Mendelson (Technion-Israel Inst. of Technology and Australian Nat. U)
Location: Mondi Seminar Room 2, Central Building
Series: Formal Sciences Seminar
Host: Jan Maas


Problems in modern statistics involve large data sets, and analyzing such sets is one of the main challenges of statistical learning theory. Over the last 15 years it has become clear that such problems have strong connections with high dimensional geometry, and the study of those connections has led to significant progress in our understanding of statistical learning theory.I will present a few of the ideas behind this geometric viewpoint to learning, and as an example, I will describe the instrumental role high dimensional geometry played in the solution of one of the classical questions in statistics - the prediction problem - with the introduction of a learning procedure that performs with the optimal sample complexity for (almost) any prediction problem.
Qr image
Download ICS Download invitation
Back to eventlist