BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Isometric Gaussian Process Latent Variable Model - Martin Jørgens
 en 
DTSTART:20211019T140000Z
DTEND:20211019T150000Z
UID:TALK164242@talks.cam.ac.uk
CONTACT:96082
DESCRIPTION:I will talk about an unsupervised model where the latent varia
 ble respects the distances and topology of the data. The ISO-GPLVM model c
 ontrols the Riemannian geometry of the data manifold. This gives the laten
 t space a stochastic distance measure. Manifold distances are modeled loca
 lly as Nakagami distributions. Stochastic distances aim to be close to the
  observed distances over a neighborhood graph. Global structure preserves 
 by the use of censoring.
LOCATION:https://cl-cam-ac-uk.zoom.us/j/94834600953?pwd=T2g4ckNwdVowdFNSRm
 4vMkZzcE5idz09
END:VEVENT
END:VCALENDAR