BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Machine Learning for Scientific Discovery\, with Examples in Fluid
  Mechanics - Steve Brunton\, U. Washington
DTSTART:20221104T160000Z
DTEND:20221104T170000Z
UID:TALK182948@talks.cam.ac.uk
CONTACT:Prof. Jerome Neufeld
DESCRIPTION:This work describes how machine learning may be used to develo
 p accurate and efficient nonlinear dynamical systems models for complex na
 tural and engineered systems.  We explore the sparse identification of non
 linear dynamics (SINDy) algorithm\, which identifies a minimal dynamical s
 ystem model that balances model complexity with accuracy\, avoiding overfi
 tting.  This approach tends to promote models that are interpretable and g
 eneralizable\, capturing the essential “physics” of the system.  We al
 so discuss the importance of learning effective coordinate systems in whic
 h the dynamics may be expected to be sparse.  This sparse modeling approac
 h will be demonstrated on a range of challenging modeling problems in flui
 d dynamics\, and we will discuss how to incorporate these models into exis
 ting model-based control efforts.  Because fluid dynamics is central to tr
 ansportation\, health\, and defense systems\, we will emphasize the import
 ance of machine learning solutions that are interpretable\, explainable\, 
 generalizable\, and that respect known physics.
LOCATION:MR2\, Centre for Mathematical Sciences\, Wilberforce Road\, Cambr
 idge
END:VEVENT
END:VCALENDAR