Study-Unit Description

Study-Unit Description


TITLE Machine Learning and Pattern Recognition Algorithms

LEVEL 05 - Postgraduate Modular Diploma or Degree Course


DEPARTMENT Communications and Computer Engineering

DESCRIPTION This study-unit describes, derives and evaluates various machine learning and pattern recognition techniques that can be later applied in various domains. This unit emphasises the theoretical and practical aspects of the pattern recognition techniques. At the end of the study-unit the student will be able to select, develop and apply machine learning algorithms to solve particular problems.

The study-unit is organized in three parts. The first part of this study-unit covers higher order features in view of non-linear models and motivates the concept using Neural Networks, followed by kernel methods and support vector machines. The feed-forward neural network is covered in depth, including the back-propagation algorithm, and an analysis of what is being learnt by the network. This part ends with generalizing regression and kernel methods to the linear basis function model.

The second part of this study-unit covers density estimation, probabilistic and Bayesian methods. It therefore starts with a review of statistics and probability. Maximum likelihood is formalized in Gaussian mixture models, and the regression problem is viewed from a probabilistic perspective, frequentist and Bayesian approach, to gain insight in the minimization of error functions and regularization. Bayesian classifiers (MAP and Naïve Bayes) are covered in this part. The Expectation Maximisation algorithm is motivated in GMM models and is derived from both a probabilistic and a Bayesian perspective.

The third part covers ensemble techniques such as Random Forests and Gradient Boosting.

Study-unit Aims:

This study-unit covers various machine learning and pattern recognition techniques that can be used in various areas. The study unit gives the student the basic statistical methodology in its various aspects for the student to be able to develop machine learning algorithms to solve particular problems.

Learning Outcomes:

1. Knowledge & Understanding
By the end of the study-unit the student will be able to:

- Distinguish between various machine learning and pattern recognition algorithms used in a number of applications;
- Analyse and demonstrate the application of a number of Pattern Recognition algorithms available including Neural Networks, Support Vector Machines, Bayesian methods and ensemble methods.

2. Skills
By the end of the study-unit the student will be able to:

- Implement the statistical representation of a problem;
- Design features and classifiers to solve a particular problem.

Main Text/s and any supplementary readings:

- Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
- Ethem Alpaydin, Introduction to Machine Learning, 3rd Edition, MIT Press, 2014.
- Kevin P. Murphy, Machine Learning; A probabilistic Perspective, MIT press, 2012.
- Tom Mitchell, Machine learning, McGraw Hill, 1997.
- R. O. Duda, P. E. Hart and D. G. Stork, Pattern Classification (2nd ed.), John Wiley and Sons, 2001. ISBN: 978-0-471-05669-0.
- Simon Haykin, Neural Networks and Learning Machines, 3rd Edition, Pearson, 2009.

ADDITIONAL NOTES Pre-requisite Qualifications: B.Sc. (Hons); B.Eng; or equivalent

STUDY-UNIT TYPE Lecture, Independent Study & Tutorial

Assessment Component/s Assessment Due Resit Availability Weighting
Assignment 1 SEM1 Yes 30%
Assignment 2 SEM1 Yes 30%
Examination (1 Hour) SEM1 Yes 40%

LECTURER/S Adrian F. Muscat
Gianluca Valentino

The University makes every effort to ensure that the published Courses Plans, Programmes of Study and Study-Unit information are complete and up-to-date at the time of publication. The University reserves the right to make changes in case errors are detected after publication.
The availability of optional units may be subject to timetabling constraints.
Units not attracting a sufficient number of registrations may be withdrawn without notice.
It should be noted that all the information in the study-unit description above applies to the academic year 2019/0, if study-unit is available during this academic year, and may be subject to change in subsequent years.