BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:UW-Madison-Physics-Events
BEGIN:VEVENT
SEQUENCE:1
UID:UW-Physics-Event-4650
DTSTART:20180202T213000Z
DURATION:PT1H0M0S
DTSTAMP:20260415T203145Z
LAST-MODIFIED:20180117T164332Z
LOCATION:2241 Chamberlin Hall
SUMMARY:Playing Newton: Learning equations of motion from data\, Physi
 cs Department Colloquium\, Ilya Nemenman\, Emory University
DESCRIPTION:Arguably\, science' goal of understanding nature can be fo
 rmulated as inferring mathematical laws that govern natural systems fr
 om experimental data. With the fast growth of power of modern computer
 s and of artificial intelligence algorithms\, there has been a recent 
 surge in attempts to automate this goal and to design\, to some extent
 \, an “artificial scientist.” I will discuss this emerging field\,
  but will focus primarily on our own approach to it. I will introduce 
 an algorithm that we have recently developed\, which allows one to inf
 er the underlying dynamical equations behind a noisy time series\, eve
 n if the dynamics are nonlinear\, and only a few of the relevant varia
 bles are measured. I will illustrate the method on applications to toy
  problems\, including inferring the iconic Newton’s law of universal
  gravitation\, as well as a few biochemical reaction networks. I will 
 end with applications to experimental biological data: modeling the la
 ndscape of possible behavioral states underlying reflexive escape from
  pain in a roundworm and (if time permits) modeling insulin secretion 
 in pancreatic beta cells.
URL:https://www.physics.wisc.edu/events/?id=4650
END:VEVENT
END:VCALENDAR