BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Scalable Gaussian Processes and the search for exoplanets
DTSTART;VALUE=DATE-TIME:20151111T130000Z
DTEND;VALUE=DATE-TIME:20151111T134500Z
DTSTAMP;VALUE=DATE-TIME:20190521T224302Z
UID:indico-contribution-939881@indico.cern.ch
DESCRIPTION:Speakers: Daniel ForemanMackey (University of Washington)\nGau
ssian Processes are a class of non-parametric models that are often used t
o model stochastic behavior in time series or spatial data. A major limita
tion for the application of these models to large datasets is the computat
ional cost. The cost of a single evaluation of the model likelihood scales
as the third power of the number of data points. In the search for transi
ting exoplanets\, the datasets of interest have tens of thousands to milli
ons of measurements with uneven sampling\, rendering naive application of
a Gaussian Process model impractical. To attack this problem\, we have dev
eloped robust approximate methods for Gaussian Process regression that can
be applied at this scale. I will describe the general problem of Gaussian
Process regression and offer several applicable use cases. Finally\, I wi
ll present our work on scaling this model to the exciting field of exoplan
et discovery and introduce a well-tested open source implementation of the
se new methods.\n\nhttps://indico.cern.ch/event/395374/contributions/93988
1/
LOCATION:CERN 222-R-001
URL:https://indico.cern.ch/event/395374/contributions/939881/
END:VEVENT
END:VCALENDAR