Isometric Gaussian Process Latent Variable Model

Add to your list(s) Download to your calendar using vCal

• Martin Jørgensen
• Tuesday 19 October 2021, 15:00-16:00
• https://cl-cam-ac-uk.zoom.us/j/94834600953?pwd=T2g4ckNwdVowdFNSRm4vMkZzcE5idz09.

If you have a question about this talk, please contact .

573203

I will talk about an unsupervised model where the latent variable respects the distances and topology of the data. The ISO -GPLVM model controls the Riemannian geometry of the data manifold. This gives the latent space a stochastic distance measure. Manifold distances are modeled locally 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.

This talk is part of the ML@CL Seminar Series series.

This talk is included in these lists:

• Hanchen DaDaDash
• ML@CL Seminar Series
• https://cl-cam-ac-uk.zoom.us/j/94834600953?pwd=T2g4ckNwdVowdFNSRm4vMkZzcE5idz09

Note that ex-directory lists are not shown.