Timezone: »
Poster
Affine Independent Variational Inference
Edward Challis · David Barber
Wed Dec 05 07:00 PM -- 12:00 AM (PST) @ Harrah’s Special Events Center 2nd Floor
We present a method for approximate inference for a broad class of non-conjugate probabilistic models. In particular, for the family of generalized linear model target densities we describe a rich class of variational approximating densities which can be best fit to the target by minimizing the Kullback-Leibler divergence. Our approach is based on using the Fourier representation which we show results in efficient and scalable inference.
Author Information
Edward Challis (University College London)
David Barber (University College London)
More from the Same Authors
-
2012 Poster: A Unifying Perspective of Parametric Policy Search Methods for Markov Decision Processes »
Thomas Furmston · David Barber -
2012 Oral: A Unifying Perspective of Parametric Policy Search Methods for Markov Decision Processes »
Thomas Furmston · David Barber -
2007 Workshop: Approximate Bayesian Inference in Continuous/Hybrid Models »
Matthias Seeger · David Barber · Neil D Lawrence · Onno Zoeter -
2006 Workshop: Advances in Models for Acoustic Processing »
David Barber -
2006 Poster: A Novel Gaussian Sum Smoother for Approximate Inference in Switching Linear Dynamical »
David Barber · Bertrand Mesot -
2006 Poster: Unified Inference for Variational Bayesian Linear Gaussian State-Space Models »
David Barber · Silvia Chiappa -
2006 Spotlight: Unified Inference for Variational Bayesian Linear Gaussian State-Space Models »
David Barber · Silvia Chiappa