BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:UW-Madison-Physics-Events
BEGIN:VEVENT
SEQUENCE:0
UID:UW-Physics-Event-7951
DTSTART:20221005T160000Z
DTEND:20221005T171500Z
DTSTAMP:20260414T154345Z
LAST-MODIFIED:20221002T153321Z
LOCATION:Online Seminar: Please sign up for our mailing list at www.ph
 ysicsmeetsml.org for zoom link
SUMMARY:Maths with transformers\, Physics ∩ ML Seminar\, François C
 harton\, Meta
DESCRIPTION:Transformers can be trained\, from synthetic data\, to sol
 ve problems of mathematics\, by considering them as translations from 
 problems into solutions. Models achieve high accuracy on a variety of 
 tasks\, learn deep mathematical properties\, and generalize out of dis
 tribution if their training set is selected with care. I illustrate th
 eir use on several problems\, from symbolic integration to numerical c
 omputation\, and symbolic regression.
URL:https://www.physics.wisc.edu/events/?id=7951
END:VEVENT
END:VCALENDAR