University of Birmingham > Talks@bham > Artificial Intelligence and Natural Computation seminars > Mixed Variational Inference

  Browse A-Z

Log in

System Admin (disabled)details Birmingham usersdetails No account?details

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

Mixed Variational Inference

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

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

Host: Prof Peter Tiňo

Abstract: The Laplace approximation has been one of the workhorses of Bayesian inference. It often delivers good approximations in practice despite the fact that it does not strictly take into account where the volume of posterior density lies. Variational approaches avoid this issue by explicitly minimising the Kullback-Leibler divergence DKL between a postulated posterior and the true (unnormalised) logarithmic posterior. However, they rely on a closed form DKL in order to update the variational parameters. To address this, stochastic versions of variational inference have been devised that approximate the intractable DKL with a Monte Carlo average. This approximation allows calculating gradients with respect to the variational parameters. However, variational methods often postulate a factorised Gaussian approximating posterior. In doing so, they sacrifice a-posteriori correlations. In this work, we propose a method that combines the Laplace approximation with the variational approach. The advantages are that we maintain: applicability on non-conjugate models, posterior correlations and a reduced number of free variational parameters. Numerical experiments demonstrate improvement over the Laplace approximation and variational inference with factorised Gaussian posteriors.

Website: https://sites.google.com/site/ngiann/

This talk is part of the Artificial Intelligence and Natural Computation seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Facts and Snacks Contemporary History Seminar Theoretical Physics Seminars

Other talks

Extending the Lax type operator for finite W-algebras TBC Quantum simulations using ultra cold ytterbium Geometry of alternating projections in metric spaces with bounded curvature Modelling uncertainty in image analysis. Sylow branching coefficients for symmetric groups
 

Talks@bham, University of Birmingham. Contact Us | Help and Documentation | Privacy and Publicity.
talks@bham is based on talks.cam from the University of Cambridge.