BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Secondmind's research activities to make Gaussian Processes indust
 ry proof - Nicolas Durrande\, Director of Research at Secondmind
DTSTART:20210331T100000Z
DTEND:20210331T113000Z
UID:TALK157384@talks.cam.ac.uk
CONTACT:Elre Oldewage
DESCRIPTION:Secondmind is developing a cloud based engine calibration tool
  based on Gaussian process models and Bayesian optimisation. In this talk 
 we present an industry point of view of the challenges associated with Gau
 ssian process models in general\, and more precisely with sparse variation
 al gaussian process models. We also present some simple modelling tips tha
 t can make a big practical difference\, and give an overview of the resear
 ch we have recently conducted to overcome what we see as the most challeng
 ing limitations of GP models. More specifically we will review some of Sec
 ondmind's contributions to make GP models more computationally efficient a
 nd more versatile.
LOCATION:https://eng-cam.zoom.us/j/82019956685?pwd=WUNSVVcrdC9IZGxQOHFhSTh
 jUjd2dz09
END:VEVENT
END:VCALENDAR