6th February 2017 - Variational Bayesian Inference: For when you look at Gibbs sampling and think “Well, that’s just too easy”. - Slides - Example Code
Variational Bayesian methods allow one to derive analytic approximations to intractable integrals arising in Bayesian inference. This talk is a beginner's introduction to using variational Bayes to approximate the posterior probability of unobserved variables in a Bayesian model, as an alternative to Monte Carlo methods.
2nd February 2017 - Beyond the console - Slides
A review of progress bars and animated plots for getting feedback during long-running R jobs, including an overview of my package longJobUtils
14th November 2016 - A guided tour of Gaussian Process Latent Variable Model (GPLVM) extensions - Slides
GPLVMs are a nonlinear extension of dual probabilistic PCA providing a generative latent variable model of high-dimensional data. After a brief introduction to GPLVMs we will present three extensions to GPLVM:
- Back-constrained GPLVM constrains the latent variables to be a function of the original data;
- Discriminative GPLVM regularizes GPLVM using Fisher's Linear Discriminant to identify features relevant to a classification task;
- Structured GPLVM is a novel technique for incorporating prior knowledge of underlying structure in the data.
10th May 2016 - Automatic Model Selection for Gaussian Processes - Slides
Gaussian processes offer a powerful framework for regression and classification tasks, with much of their flexibility being due to the dizzying array of possible kernels one can use to define the types of structure they model. In this talk we discuss methods of automatically discovering a suitable kernel for a given task using sequential model selection.
26th January 2016 - Bayesian Model Selection - Slides
A brief overview of Bayesian model selection.