BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Inter-domain Deep Gaussian Processes - Tim G. J. Rudner\, Universi
 ty of Oxford
DTSTART:20201016T120000Z
DTEND:20201016T140000Z
UID:TALK151939@talks.cam.ac.uk
CONTACT:Francisco Vargas
DESCRIPTION:*1-on-1 Meetings with the Speaker*:\n\nThe talk is scheduled f
 or one hour and will be followed by one hour of 1-on-1 meetings with the s
 peaker. To book a 1-on-1 slot with the speaker\, email the organizer befor
 e the event.\n\n\n*Paper:*\n\n"Inter-domain Deep Gaussian Processes":http:
 //timrudner.com/papers/Inter-domain_Deep_Gaussian_Processes/Rudner2020_Int
 er-domain_Deep_Gaussian_Processes.pdf (ICML 2020)\n\n*Abstract:*\n\nInter-
 domain Gaussian processes (GPs) allow for high flexibility and low computa
 tional cost when performing approximate inference in GP models. They are p
 articularly suitable for modeling data exhibiting global structure but are
  limited to stationary covariance functions and thus fail to model non-sta
 tionary data effectively. We propose Inter-domain Deep Gaussian Processes\
 , an extension of inter-domain shallow GPs that combines the advantages of
  inter-domain and deep Gaussian processes (DGPs)\, and demonstrate how to 
 leverage existing approximate inference methods to perform simple and scal
 able approximate inference using inter-domain features in DGPs. We assess 
 the performance of our method on a range of regression tasks and demonstra
 te that it outperforms inter-domain shallow GPs and conventional DGPs on c
 hallenging large-scale real-world datasets exhibiting both global structur
 e as well as a high-degree of non-stationarity.\n\n*Keywords:* Gaussian Pr
 ocesses\, Variational Inference\, Bayesian Deep Learning\n\n\n*About the S
 peaker:*\n\nTim G. J. Rudner is a PhD Candidate in the Department of Compu
 ter Science at the University of Oxford\, supervised by Yarin Gal and Yee 
 Whye Teh. His research interests span Bayesian deep learning\, reinforceme
 nt learning\, and variational inference. He holds a master’s degree in s
 tatistics from the University of Oxford and an undergraduate degree in mat
 hematics and economics from Yale University. Tim is also an AI Fellow at G
 eorgetown University's Center for Security and Emerging Technology (CSET)\
 , a Fellow of the German National Academic Foundation\, and a Rhodes Schol
 ar.\n\n*Website:* "http://timrudner.com":http://timrudner.com\n\n\n\nThis 
 talk is part of the ML@CL Seminar Series with a focus on early career rese
 archers and topics relevant to machine learning and statistics.
LOCATION:https://dtudk.zoom.us/j/63696188914?pwd=L3RNZFlVWjdvdCtwMTAzaXA0U
 HVlQT09
END:VEVENT
END:VCALENDAR