Thang Bui

icon indicating a website link thangbui.github.io

icon company Australian National University icon location Canberra, Australia icon location Lecturer in Machine Learning @ Australian National University, AU.

Results 4 comments of Thang Bui

Approximate Power-EP for Deep GPs with intermediate variables for density models/latent variable models

Hi @wil-j-wil, Unfortunately, there are still many models missing or not fully implemented. My goal originally was to have all of the following models + inference schemes under one package...

Quadrature/Monte Carlo

Notes: - use logsumexp - use the reparameterisation trick/store random numbers/only change them after t iterations - high-dimensional quadratures?

Quadrature/Monte Carlo

Notes: * Simple Monte Carlo for the latent variables should work for GPLVMs, GPSSMs and deep GPs with hidden variables. This should be enough because, conditioned on a latent sample,...

Frigola et al's VI for GPSSMs

The idea is to use parameterised q(x) (Markov or mean field) instead of using SMC.