University of Cambridge > Talks.cam > Machine Learning Reading Group @ CUED > Automated Augmented Conjugate Inference for Gaussian Processes

  Browse A-Z

Log in

Googledetails Microsoftdetails

🔐 Log In

Information on

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

ℹ️ Information

Automated Augmented Conjugate Inference for Gaussian Processes

Download to your calendar using vCal

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

Gaussian Processes are a tool of choice for modelling function with uncertainties. However, inference is only tractable analytically for the classical case of regression with Gaussian noise since all other likelihoods are not conjugate with the Gaussian prior.

In this talk I will show how one can transform a large class of likelihoods into conditional conjugate distributions by augmenting them with latent variables. These augmented models have the advantage that, while the posterior inference is still not fully analytic, the full conditionals are! Consequently, one can work easily (and efficiently!) with algorithms like Gibbs sampling or Coordinate Ascent VI (CAVI) and outperform existing inference methods.

Reference:

Galy-Fajou, Théo, Florian Wenzel, and Manfred Opper. “Automated Augmented Conjugate Inference for Non-conjugate Gaussian Process Models.” International Conference on Artificial Intelligence and Statistics. PMLR , 2020. https://arxiv.org/abs/2002.11451

This talk is part of the Machine Learning Reading Group @ CUED series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

History Faculty Board Room Perham Seminar Room, Sanger Building SS03, William Gates Building

Other talks

CSaP Annual Cleevely Lecture 2026 Polar Oceans Seminar Talk Morning Break Title TBC Group Work What’s my reproduction number? Life history strategies: an interdisciplinary perspective
 

© 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity