Timezone: »
We introduce a variational Bayesian inference algorithm which can be widely applied to sparse linear models. The algorithm is based on the spike and slab prior which, from a Bayesian perspective, is the golden standard for sparse inference. We apply the method to a general multi-task and multiple kernel learning model in which a common set of Gaussian process functions is linearly combined with task-specific sparse weights, thus inducing relation between tasks. This model unifies several sparse linear models, such as generalized linear models, sparse factor analysis and matrix factorization with missing values, so that the variational algorithm can be applied to all these cases. We demonstrate our approach in multi-output Gaussian process regression, multi-class classification, image processing applications and collaborative filtering.
Author Information
Michalis Titsias (Athens University of Economics and Business)
Miguel Lazaro-Gredilla (DeepMind)
More from the Same Authors
-
2021 Poster: Perturb-and-max-product: Sampling and learning in discrete energy-based models »
Miguel Lazaro-Gredilla · Antoine Dedieu · Dileep George -
2015 Poster: Inference for determinantal point processes without spectral knowledge »
Rémi Bardenet · Michalis Titsias -
2013 Poster: Variational Inference for Mahalanobis Distance Metrics in Gaussian Process Regression »
Michalis Titsias · Miguel Lazaro-Gredilla -
2012 Poster: Bayesian Warped Gaussian Processes »
Miguel Lazaro-Gredilla -
2011 Poster: Variational Gaussian Process Dynamical Systems »
Andreas Damianou · Michalis Titsias · Neil D Lawrence -
2009 Poster: Inter-domain Gaussian Processes for Sparse Inference using Inducing Features »
Miguel Lazaro-Gredilla · Anibal R Figueiras-Vidal -
2008 Poster: Efficient Sampling for Gaussian Process Inference using Control Variables »
Michalis Titsias · Neil D Lawrence · Magnus Rattray -
2008 Spotlight: Efficient Sampling for Gaussian Process Inference using Control Variables »
Michalis Titsias · Neil D Lawrence · Magnus Rattray -
2007 Poster: The Infinite Gamma-Poisson Feature Model »
Michalis Titsias