Module 02640 (2004)

Syllabus page 2004/2005

06-02640
Machine Learning

Level 2/I

Ata Kaban
10 credits in Semester 2

Links | Outline | Aims | Outcomes | Prerequisites | Teaching | Assessment | Books | Detailed Syllabus


The Module Description is a strict subset of this Syllabus Page. (The University module description has not yet been checked against the School's.)

Changes and updates

Assessment changed: 20% continuous assessment added.


Relevant Links


Outline

The module will provide a good foundation in machine learning. It will compare and contrast human learning with machine learning. It will examine the limitations of machine learning, the role of hypothesis bias and hypothesis representation.


Aims

The aims of this module are to:

  • introduce the basic concepts and terminology of machine learning
  • give an overview on the main approaches to machine learning (except neural nets, dealt with in a separate module)
  • show similarities and differences between different approaches
  • present basic principles for the classification of approaches to machine learning
  • give practical experience of applying machine learning algorithms to classification and data analysis problems

Learning Outcomes

On successful completion of this module, the student should be able to: Assessed by:
1demonstrate a knowledge and understanding of the main approaches to machine learning Examination
2demonstrate the ability to apply the main approaches to unseen examples Examination, Continuous Assessment
3demonstrate an understanding of the differences, advantages and problems of the main approaches in machine learning Examination, Continuous Assessment
4demonstrate an understanding of the relationship between machine learning and human learning Examination
5demonstrate an understanding of the main limitations of current approaches to machine learning, and be able to discuss possible extensions to overcome these limitations Examination, Continuous Assessment
6demonstrate a practical understanding of the use of machine learning algorithms Continuous Assessment

Restrictions, Prerequisites and Corequisites

Restrictions:

None

Prerequisites:

06-18188 (Introduction to AI) (or equivalent)

Co-requisites:

None


Teaching

Teaching Methods:

22 hrs lectures & exercise and lab classes

Contact Hours:

22


Assessment

  • Supplementary (where allowed): As the sessional assessment
  • 2 hr examination (80%), continuous assessment (20%). Resit by examination only.

Recommended Books

TitleAuthor(s)Publisher, Date
Machine LearningT Mitchell
Artificial Intelligence, a Modern ApproachRussell S & Norvig P1995
Pattern ClassificationR.O. Duda, P.E. Hart & D.G.Stork2000
Reinforcement Learning: an IntroductionR Sutton & A. Barto
Modelling the Internet and the Web Pierre Baldi, Paolo Frasconi, and Padhraic Smyth 2003
Support Vector Machines and Other Kernel-Based Learning MethodsNello Cristianini & John Shawe-Taylor2000

Detailed Syllabus

  1. Basic notions of learning, introduction to learning algorithms, literature
  2. Probabilistic Sequence models. Markov Models, Aggregate Markov Models, Basic notions of Hidden Markov Models
  3. Reinforcement Learning I (basic concepts, episodic and non-episodic tasks, reward function, Markov decision processes, policy, state value, Bellman's equations), Q-learning in deterministic environments
  4. Reinforcement Learning II (finding best policy, Dynamic Programming, Q learning in non-deterministic environments
  5. Unsupervised Learning. Dimensionality reduction. Latent Semantic Analysis
  6. Independent Component Analysis. The coctail party problem.
  7. Bayesian classification and clustering
  8. Visualisation and exploratory data analysis
  9. Decision Tree Learning
  10. Support Vector Machines and Kernel Machines
  11. Instance-Based Learning (k-nearest neighbour, Case-based Reasoning)
  12. Revision

Last updated: 21 Jan 2005

Source file: /internal/modules/COMSCI/2004/xml/02640.xml

Links | Outline | Aims | Outcomes | Prerequisites | Teaching | Assessment | Books | Detailed Syllabus