Machine Learning - 2019 entry

Machine learning has emerged mainly from computer science and artificial intelligence, and draws on methods from a variety of related subjects including statistics, applied mathematics and more specialized fields, such as pattern recognition and neural computation. Applications are, for example, image and speech analysis, medical imaging, bioinformatics and exploratory data analysis in natural science and engineering. This module will provide you with a thorough grounding in the theory and application of machine learning, pattern recognition, classification, categorisation, and concept acquisition.  Hence, it is particularly suitable for Computer Science, Mathematics and Engineering students and any students with some experience in probability and programming.

PRE-REQUISITE MODULES ECM3420 or ECMM445

In this data-driven era, modern technologies are generating massive and high-dimensional datasets. This module aims to give you an understanding of computational methods used in modern data analysis.

In particular, this module aims to impart knowledge and understanding of machine learning methods from basic pattern-analysis methods to state-of-the-art research topics; to give you experience of data-modelling development in practical workshops. Neural Networks, Bayesian methods and kernel-based algorithms will be introduced for extracting knowledge from large data sets of patterns (data mining techniques) where it is important to have explicit rules governing machine learning and pattern recognition. Recent development of techniques and algorithms for big-data analysis will also be addressed.

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

Module Specific Skills and Knowledge:

1. apply advanced and complex principles for statistical machine learning to various data analysis;

2. analyse novel pattern recognition and classification problems; establish statistical models for them and write software to solve them;

3. apply a range of supervised and unsupervised machine learning techniques to a wide range of real-life applications.

Discipline Specific Skills and Knowledge:

​4. state the importance and difficulty of establishing a principled probabilistic model for pattern recognition;

5. apply a number of complex and advanced mathematical and numerical techniques to a wide range of problems and domains.

Personal and Key Transferable / Employment Skills and Knowledge:

6. identify the compromises and trade-offs which must be made when translating theory into practice;

7. critically read and report on research papers;

8. conduct small individual research projects.

- Introductory material: Practical motivation for machine learning, basic ideas of supervised and unsupervised learning, classification, regression;

- Describing data;

- Latent descriptions: k-means, maximum likelihood; mixture models; PCA; ICA;

- Unsupervised learning: Clustering; Locality Sensitive Hashing;

- Supervised models: k-nearest neighbours, linear and non-linear regression, linear discriminant analysis, logistic regression, SVM and maximum margin classifiers;

- Loss functions and maximum likelihood estimators;

- Bayesian learning & sampling;

- Neural nets and deep learning;

- Evaluation of performance, dataset balance;

- Ensemble methods: boosting, bagging, decision trees and random forests Metric learning;

- Markov decision processes: Reinforcement learning (Q-learning).

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: