SoftwareThis software is freely available for academic use only.
- GLOMO: MATLAB Toolbox for learning layers from video.
- MATLAB software on sampling in Gaussian processes using control variables . Control points algorithm described in the paper: Efficient Sampling for Gaussian Process Inference using Control Variables, NIPS 2009. The software requires the minimize.m function of C. Rasmussen.
- First version of my software on sparse variational GPs. The method described in the paper: Variational Learning of Inducing Variables in Sparse Gaussian Processes. AISTATS, 2009.
- Code for variational sparse linear models. Based on the paper: Spike and Slab Variational Inference for Multi-Task and Multiple Kernel Learning, NIPS, 2012. It can deal with: sparse linear regression; sparse factor analysis and PCA; multi-output Gaussian process regression and more.
- Code for Variational Inference over kernel GP hyperparameters . Based on the paper: Variational Inference for Mahalanobis Distance Metrics in Gaussian Process Regression, NIPS, 2013.
- Code for doubly stochastic variational inference. Based on the paper: Doubly Stochastic Variational Bayes for non-Conjugate Inference, ICML, 2014. Examples on automatic variable/feature selection in Bayesian logistic regression and Gaussian process hyperparameters inference are included.
- Code applying local expectation gradients to sigmoid belief nets . Based on the paper: Local Expectation Gradients for Black Box Variational Inference, NIPS, 2015. Examples for learning sigmoid belief nets on MNIST are included.
- Variational GP-LVM code . (written together with Andreas C. Damianou and Neil D. Lawrence)