BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:UW-Madison-Physics-Events
BEGIN:VEVENT
SEQUENCE:1
UID:UW-Physics-Event-6428
DTSTART:20210505T160000Z
DTEND:20210505T171500Z
DTSTAMP:20260414T221256Z
LAST-MODIFIED:20210506T004407Z
LOCATION:Online Seminar: Please sign up for our mailing list at www.ph
 ysicsmeetsml.org for zoom link
SUMMARY:Explorations at the Physics ∩ ML Interface\, Physics ∩ ML 
 Seminar\, Kyle Cranmer\, New York University
DESCRIPTION:Instead of focusing on a specific application\, I will dis
 cuss a few projects that explore the Physics ∩ ML Interface. How do 
 we incorporate our physical insight into the underlying causal mechani
 sm into the inductive bias of machine learning architectures? Is that 
 helpful or necessary? Why do we care if a model is interpretable? Wher
 e do we stand on the spectrum between ML-supercharged data analysis an
 d an AI / robot scientist? How does this line of thinking influence re
 search in AI and ML?
URL:https://www.physics.wisc.edu/events/?id=6428
END:VEVENT
END:VCALENDAR