Gaussian Processes with Non-Gaussian Likelihoods
Tutorial, University of Warwick, Coventry, UK
In this tutorial, I gave a quick overview of some inference methods for Gaussian Processes with non-Gaussian likelihoods. The slides can be found here.
Talks and Presentations
Gaussian Process Models for Multi-Task Learning
Talk, University of Warwick, Coventry, UK
Multi-task learning refers to a framework in machine learning where multiple tasks sharing a common domain are learned simultaneously. In this talk, I explained how to incorporate this framework within Bayesian modelling using Gaussian process models. The talk describes the general structure of a multi-output GP, and explores some of the most common kernel structures that correspond to such models. The sildes are available here.
Approximate Bayesian Computation Approximately!
Talk, YHA Ravenstor, Peak District, UK
Light hearted talk on Approximate Bayesian Computation (ABC) given at the annual retreat of my CDT. The talk can be viewed here.