This is a series of six half-day workshops delivered online. The course will help researchers develop skills to understand complex data sets by analysing data using multivariate methods. This is a practical course that uses Bayesian principles of model selection, regularisation and inference. Each session is split into lectures and practicial data analysis sessions in which you'll run through WorkSheets. You'll need Matlab installed on your computer and have some familiarity with Matlab coding. This brief Matlab Course can get you up to speed.
Introduction to the PEARL-Neuro data set
Introduction to Bayesian inference
Perception as Inference
Zip file for session1
Video Recordings:
(1A) Intro to Data Set 1 (1B) Intro to Bayesian Inference (1C) Intro to Data Set 2
Optimal Fusion of Multivariate Data
Multiple Linear Regression - where you have multiple independent variables (‘inputs’) and one dependent variable (‘output’)
Eigenvectors, Determinants and the Multivariate Gaussian
Bayesian Model Comparison for Linear Models and Principal Component Analysis
Zip file for session2
Bayesian Data Analysis software: see Session 3
Video Recordings:
(2A) Optimal Fusion , (2B) Bayesian GLM , (2C) Not Recorded (The Multivarate Gaussian), (2D) GLM model comparison and PCA , (2E) PCA Model Comparison
Where you have multiple independent variables (‘inputs’) and one dependent variable that is binary (‘output’)
Mass Univariate Analysis
Logistic Regression and Bayesian Logistic Regression
Support Vector Machines
Sparse Bayesian Models
Bayesian Data Analysis software: BDA
Zip file for session3
Video Recordings:
(3A) Mass Univariate Analyses , (3B) Logistic Regression , (3C) Support Vector Machines , (3D) Sparse Bayesian Modelling
How different are two probability distributions?
Variational Bayes
Canonical Correlaton Analysis
Zip file for session4
Includes Mixtures of CVAs, Mixtures of Gaussians and Neural Networks (Multilayer Perceptrons).
Some encoding/decoding problems can only be addressed using nonlinear methods.