Learning from Data - 2019 entry

Artificially intelligent machines and software must assimilate data from their environment and make decisions based upon it. Likewise, we live in a data-rich society and must be able to make sense of complex datasets. This module will introduce you to machine learning methods for learning from data. You will learn about the principal learning paradigms from a theoretical point of view and gain practical experience through a series of workshops. Throughout the module, there will be an emphasis on dealing with real data, and you will use, modify and write software to implement learning algorithms. It is often useful to be able to visualise data and you will gain experience of methods of reducing the dimension of large datasets to facilitate visualisation and understanding.

The module will also cover some recent neural network architectures and related learning algorithms.

This module aims to equip you with the fundamentals of machine learning and at the same time discuss technical aspects of some well-known machine learning models and related learning algorithms. It will provide a thorough grounding in the theory and application of machine learning and statistical techniques for classification, regression and unsupervised methods (clustering and dimension reduction). The module will cover kernel methods and neural networks (feed-forward architectures only).

On successful completion of this module you should be able to:

Module Specific Skills and Knowledge

Discipline Specific Skills and Knowledge

Personal and Key Transferable / Employment Skills and Knowledge

- introductory material: basic ideas of classification, regression and preliminary concepts;

- data description;

- statistical preliminaries: probability, standard distributions and densities, Bayes rule;

- modelling and learning: models, noise, maximum likelihood learning and error functions; generalisation;

- classification: decision boundaries; k-nn classifier, linear discriminants;

- receiver operating characteristics: loss functions; ROC curves and their optimisation;

- optimisation for learning;

- neural networks, multi-layer feed-forward networks, RBF, back-propagation learning algorithm, auto-encoders;

- kernel-based classifiers, kernel trick;

- feature extraction: sequential forwards/backwards selection;

- unsupervised methods: clustering,  k-means, kernel k-means, cluster validation;

- dimension reduction and visualisation: PCA, ICA, linear and nonlinear methods for visualisation, MDS and isomap.

Two workshops will be assessed. Other workshops will have formative components.

MSc students will also develop a practical project with formative feedback available. Students on the level-3 version of this module (ECM3420) are not requested to develop a project. Accordingly, for students on ECM3420, the two assessed courseworks will be worth 20% each (40% of the total mark).

Referred and deferred assessment will normally be by examination. For referrals, only the examination will count, a mark of 50% being awarded if the examination is passed. For deferrals, candidates will be awarded the higher of the deferred examination mark or the deferred examination mark combined with the original coursework mark.

Basic reading:

ELE: http://vle.exeter.ac.uk/

Web based and Electronic Resources:

Other Resources:

Reading list for this module: