University of Surrey - Guildford

Registry > Module Catalogue
View Module List by A.O.U. and Level  Alphabetical Module Code List  Alphabetical Module Title List  Alphabetical Old Short Name List  View Menu 
2010/1 Module Catalogue
 Module Code: EEE3011 Module Title: MACHINE INTELLIGENCE
Module Provider: Electronic Engineering Short Name: EE3.MI
Level: HE3 Module Co-ordinator: KITTLER JV Prof (Elec Eng)
Number of credits: 15 Number of ECTS credits: 7.5
Module Availability

Spring semester

Assessment Pattern

Unit(s) of Assessment
Weighting Towards Module Mark( %)
Written examination: Closed Book, 2 hours, 3 questions out of 5
Assignments: Individual Pattern Recognition System Design and Testing, Report 6-10 pages

Module Overview



Module Aims

To give the option to students wishing to specialise in Machine Intelligence with focus on pattern recognition and artificial intelligence.

Learning Outcomes

Students should understand and assimilate terminology and concepts in artificial intelligence and sensory data perception introduced in the module.  They should understand how these concepts are expressed mathematically and should be able to manipulate mathematical models to solve problems and predict effects.  They should gain appreciation of the relevance of the material in the engineering and applied context.

Module Content
(12 hours)    (KW)
[1-2]     Historical Overview:    Definition of Artificial Intelligence (AI). Application areas. General problem solving vs specific knowledge. Complexity.
[3-6]     Heuristic Search:     Uninformed vs informed search strategies. Formal properties A*. Minimax game search, alpha-beta pruning.   
[7-10]   Logic and Resolution:     Knowledge Representation, Propositional and Predicate Calculus. Inference Rules. Clause Form. Resolution Strategies. 
[11-12] Case studies:    Applications in medicine.
PATTERN RECOGNITION:    (12 hours)     (JVK)
[1]           Introduction. The pattern recognition problem. Model of pattern recognition system. Statistical and structural approach. Basic terminology. 
[2-3]        Decision Theory.  Review of mathematical prerequisites and probability theory. Elements of statistical decision theory. Bayes decision rule. Parametric decision rule for normally distributed classes.
[4-5]        Classifiers.    Discriminant functions. Support Vector Machines. Nearest neighbour decision rules. Error estimation. Multiple classifiers.
[6-7]        Non-supervised Learning. Cluster analysis. Concept of a cluster. Cluster models. Similarity measures. Clustering criteria and algorithms.
[8-9]        Feature Selection and Extraction. Class separability measures. Feature set search algorithms (branch and bound, sub-optimal algorithms). Feature extraction. Karhunen-Loeve transform.
[10-12]    Case Studies.  One of the following: Face recognition. Detection of microcalcification in digital mammograms. Video segmentation. Speech recognition.
NEURAL NETWORKS:                     (6 hours)         (TW)
[1]           Artificial NN models: biological plausibility, feedback concepts, network architecture overview, discriminate functions, learning paradigms, linear associative memories.
[2-3]        Single-Layer feed forward: discrete and continuous perceptrons, convergence theorem, comparison with Gaussian classifier, nonseparability, LMS algorithms, steepest descent.
[4-5]        Multilayer Perceptrons: Learning in multilayer perceptrons, target function, objective functions, parameters of learning process, convergence, network growing and pruning.

[6]           Case studies: Applications of neural network techniques.

Methods of Teaching/Learning
Lectures:                 Supported by example classes
Assignments:        Pattern recognition system design and testing  

Example classes: Timetabled example classes and demonstrating delivered by RA/PhD student

Selected Texts/Journals
Recommended by Professor Kittler
Webb, A     Statistical Pattern Recognition     0340741643     Arnold                               B
Recommended by Dr K Wells
Russel & Norvig     Artificial Intelligence: A   0-13-360124-2     Prentice-Hall                   B
Recommended by Dr T Windeatt
Haykin, S     Neural Networks 2nd Ed.   0-13-273350-1   Prentice-Hall                             B

Bishop,C     Neural Networks for Pattern Recognition 0-19-853864-2     Oxford             B

Last Updated

31 July 2009