Print View: COMP4702 - Sem 1 2008 - St Lucia

1. General Course Information

1.1 Course Details

Course Code: COMP4702 Course Title: Machine Learning
Coordinating Unit: School of Information Technology and Electrical Engineering
Semester: Semester 1, 2008 Mode: Internal
Level: Undergraduate
Location: St Lucia
Number of Units: 2 Contact Hours Per Week: 2L3C
Pre-Requisites: MATH1051
Recommended Pre-Requisites: COMP3702
Incompatible: COGS4021 or COMP3700 or COMP7703 or ELEC4700 or ELEC7701

Course Description: Machine learning is a branch of artificial intelligence concerned with the development & application of adaptive algorithms that use example data or previous experience to solve a given problem. Topics include: learning problems (e.g regression, classification, unsupervised, reinforcement) & theory, neural networks, statistical & probabilistic models, clustering, ensembles, implementation issues, applications (e.g. bioinformatics, cognitive science, forecasting, robotics, signal & image processing).

Assumed Background: Pre: (COMP3701, COMP3702, COMP2702 or COGS2000) and intro calculus or 3 years of engineering study.
Inc: COMP3700/ELEC4700/ELEC7701 and COGS4021.
The course will deal with concepts using algorithms and data structures, mathematics, statistics and probability appropriate for a 4th year engineering, information technology or science student.

1.3 Course Staff

Course Coordinator: Dr Marcus Gallagher
Phone: 3365 6197 Email: marcusg@itee.uq.edu.au Homepage: www.itee.uq.edu.au/~marcusg
Campus: St Lucia Building: Axon Building (Map) Room: 502
Consultation: Arrange an appointment via email.

1.4 Timetable

Timetables are available on mySI-net.

Additional Timetable Information
Note: Practical and Tutorial sessions will only be held in specified weeks during the course.

2. Aims, Objectives & Graduate Attributes

2.1 Course Aims

Students will gain a fundamental understanding of a wide range of Machine Learning techniques and algorithms, including supervised and unsupervised learning. Students will also gain an appreciation for the practical applications of machine learning techniques, and will gain experience in implementation and using machine learning methods.

2.2 Learning Objectives

After successfully completing this course you should be able to:

1  Implement and apply machine learning techniques to solve problems
2  Appreciate machine learning research ideas (state-of-the-art) and practice (journal and conference publication, peer review).
3  Understand the main issues and core problems that the field of machine learning is concerned with.
4  Understand the relationships between machine learning and other fields (artificial intelligence, statistics, pattern recognition, optimization).
5  Understand the main theoretical and conceptual issues in machine learning.
6  Understand and be able to apply probabilistic (generative) supervised machine learning models.
7  Understand and be able to apply discriminative supervised machine learning models.
8  Understand and be able to apply techniques for density estimation, dimensionality reduction, clustering and learning sequences.

2.3. Graduate Attributes

Successfully completing this course will contribute to the recognition of your attainment of the following UQ (Undergrad Pass) graduate attributes:

GRADUATE ATTRIBUTE	LEARNING OBJECTIVES
*A. IN-DEPTH KNOWLEDGE OF THE FIELD OF STUDY*
A1. A comprehensive and well-founded knowledge in the field of study.	1, 3, 4, 5, 6, 7, 8
A4. An understanding of how other disciplines relate to the field of study.
A5. An international perspective on the field of study.
*B. EFFECTIVE COMMUNICATION*
B1. The ability to collect, analyse and organise information and ideas and to convey those ideas clearly and fluently, in both written and spoken forms.	3, 5, 6, 7, 8
B2. The ability to interact effectively with others in order to work towards a common outcome.
B3. The ability to select and use the appropriate level, style and means of communication.
B4. The ability to engage effectively and appropriately with information and communication technologies.	1, 2, 5, 6, 7, 8
*C. INDEPENDENCE AND CREATIVITY*
C1. The ability to work and learn independently.	1, 2, 3, 4, 6, 7, 8
C3. The ability to generate ideas and adapt innovatively to changing environments.	8
C4. The ability to identify problems, create solutions, innovate and improve current practices.	1, 6, 7, 8
*D. CRITICAL JUDGEMENT*
D1. The ability to define and analyse problems.	1, 5, 6, 7, 8
D2. The ability to apply critical reasoning to issues through independent thought and informed judgement.	2, 5, 6, 8
D3. The ability to evaluate opinions, make decisions and to reflect critically on the justifications for decisions.	1, 2, 3, 7, 8
*E. ETHICAL AND SOCIAL UNDERSTANDING*
E1. An understanding of social and civic responsibility.
E2. An appreciation of the philosophical and social contexts of a discipline.	2, 5
E4. A knowledge and respect of ethics and ethical standards in relation to a major area of study.
E5. A knowledge of other cultures and times and an appreciation of cultural diversity.

Successfully completing this course will contribute to the recognition of your attainment of the following Engineers Australia graduate attributes:

GRADUATE ATTRIBUTE	LEARNING OBJECTIVES
1. Ability to apply knowledge of basic science and engineering fundamentals	1, 5, 6, 7, 8
2. Ability to communicate effectively, not only with engineers, but also with the community at large	8
3. In-depth technical competence in at least one engineering discipline	3, 6, 7, 8
4. Ability to undertake problem identification, formulation and solution	1, 3, 6, 7, 8
5. Ability to utilise a systems approach to design and operational performance
6. Ability to function effectively as an individual and in multi-disciplinary and multi-cultural teams, with the capacity to be a team leader or manager as well as an effective team member
7. Understanding of the social, cultural, global and environmental responsibilities of the professional engineer, and for the need for sustainable development
8. Understanding of the principles of sustainable design and development
9. Understanding of and commitment to professional and ethical responsibilities
10. Expectation and capacity to undertake life-long learning

3. Learning Resources

3.1 Required Resources

Alpaydin, E. Introduction to Machine Learning. MIT Press, 2004

3.2 Recommended Resources

R. Duda, P. Hart and D. Stork. Pattern Classification, Second edition. Wiley, 2001.

Bishop, C. M. Pattern Recognition and Machine Learning. Springer. 2006

T. Hastie, R. Tibshirani and J. Friedman. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer. 2001.

D. Hand, H. Mannila and P. Smyth, Principles of Data Mining, MIT Press, 2001.

T. M. Mitchell, Machine Learning, McGraw-Hill, 1997

3.3 University Learning Resources

Access to required and recommended resources, plus past central exam papers, is available at the UQ Library website (http://library.uq.edu.au/search/r?SEARCH=COMP4702).

The University offers a range of resources and services to support student learning. Details are available on the myServices website (https://student.my.uq.edu.au/).

3.4 School of Information Technology and Electrical Engineering Learning Resources

Students enrolled at St Lucia who wish to retain a hard copy of this profile can use the free print quota provided each semester to students enrolled in courses in the School of Information Technology & Electrical Engineering. For information on how to use this print quota, see the School Policy on Student Photocopying and Printing (St Lucia) (http://www.itee.uq.edu.au/about_ITEE/policies/copy-print.html). Students enrolled at the Ipswich campus will either be provided with a hard copy or given directions in class on how to obtain a free copy.

ITEE course websites can be found at http://www.itee.uq.edu.au/~COURSECODE. Many ITEE courses also have Usenet newsgroups, named uq.itee.COURSECODE. Instructions for accessing newsgroups are available at http://studenthelp.itee.uq.edu.au/faq/1stYearFAQ.html#accessnews.

3.5 Other Learning Resources & Information

Additional resources (e.g journal and conference papers, web sites) may be referred to during classes.

4. Teaching & Learning Activities

4.1 Learning Activities

Date	Lecture	Practical
25 Feb - 2 Mar	Thu 12:00 Introduction; Supervised Learning Learning Objectives: 3, 4, 5 Readings/Ref: Alpaydin (Chapters 1 and 2);
3 Mar - 9 Mar	Thu 12:00 Bayesian Decision Theory Learning Objectives: 2, 3, 5 Readings/Ref: Alpaydin (Chapter 3);	Wed 14:00 Supervised Learning Learning Objectives: 1, 3, 4, 5 Readings/Ref: Alpaydin ;
10 Mar - 16 Mar	Thu 12:00 Parametric Methods Learning Objectives: 3, 5, 6 Readings/Ref: Alpaydin (Chapters 4 & 5);	Wed 14:00 Bayesian Decision Theory Learning Objectives: 1, 2, 3, 5, 6 Readings/Ref: Alpaydin ;
17 Mar - 23 Mar	Thu 12:00 Dimensionality Reduction Learning Objectives: 3, 8 Readings/Ref: Alpaydin (Chapter 6);	Wed 14:00 Parametric Methods Learning Objectives: 1, 3, 5, 6 Readings/Ref: Alpaydin ;
31 Mar - 6 Apr	Thu 12:00 Clustering Learning Objectives: 3, 8 Readings/Ref: Alpaydin (Chapter 7);	Wed 14:00 Dimensionality Reduction Learning Objectives: 1, 3, 8 Readings/Ref: Alpaydin ;
7 Apr - 13 Apr	Thu 12:00 Nonparametric Methods Learning Objectives: 3, 6 Readings/Ref: Alpaydin (Chapter 8);	Wed 14:00 Clustering Learning Objectives: 1, 3, 8 Readings/Ref: Alpaydin ; Bishop06
14 Apr - 20 Apr	Thu 12:00 Linear Discrimination Learning Objectives: 1, 5, 7 Readings/Ref: Alpaydin (Chapter 10);	Wed 14:00 Nonparametric Methods Learning Objectives: 1, 3, 6 Readings/Ref: Alpaydin ;
21 Apr - 27 Apr	Thu 12:00 Multilayer Perceptrons Learning Objectives: 3, 6 Readings/Ref: Alpaydin (Chapter 11);	Wed 14:00 Linear Discrimination Learning Objectives: 1, 5, 7 Readings/Ref: Alpaydin ;
28 Apr - 4 May	Thu 12:00 Local Models Learning Objectives: 2, 3, 6, 7 Readings/Ref: Alpaydin (Chapter 12);	Wed 14:00 Multilayer Perceptrons Learning Objectives: 1, 3, 4, 5, 7 Readings/Ref: Alpaydin ;
5 May - 11 May	Thu 12:00 Hidden Markov Models Learning Objectives: 3, 5, 8 Readings/Ref: Alpaydin (Chapter 13);	Wed 14:00 Local Models Learning Objectives: 1, 2, 3, 6, 7 Readings/Ref: Alpaydin ;
12 May - 18 May	Thu 12:00 Combining Multiple Learners Learning Objectives: 3, 5 Readings/Ref: Alpaydin (Chapter 15);
19 May - 25 May	Thu 12:00 Assessing and Comparing Learning Objectives: 3, 5 Readings/Ref: Alpaydin (Chapter 14);
26 May - 1 Jun	Thu 12:00 Course Review; Exam Preview Learning Objectives: 3, 5 Readings/Ref: Alpaydin ;

4.2 Other Teaching and Learning Activities Information

You are not required to attend any of the teaching sessions (except those in which an assessment activity is taking place), however, you are strongly encouraged to do so. The lectures, tutorials and pracs have been specifically designed to aid your learning of the course material. Failure to attend a session may result in you being disadvantaged. It is up to you to find out what happened at any class session that you miss.

5. Assessment

5.1 Assessment Summary

This is a summary of the assessment in the course. For detailed information on each assessment, see 5.5 Assessment Detail below.

Assessment Task	Due Date	Weighting	Learning Objectives
Problem Set/s Assignment 1	11 Apr 08	20%	1, 2, 3, 4
Problem Set/s Assignment 2	16 May 08	20%	1, 2, 3, 4
Exam - during Exam Period (Central) Final Exam	Examination Period	60%	3, 4, 5, 6, 7, 8

5.2 Course Grading

Grade 1, Fail: Fails to demonstrate most or all of the basic requirements of the course: This grade will be awarded when the combined total marks from all assessment is at least 0% but less than 20% of total marks available.

The minimum percentage required for a grade of 1 is: 0%

Grade 2, Fail: Demonstrates clear deficiencies in understanding and applying fundamental concepts; communicates information or ideas in ways that are frequently incomplete or confusing and give little attention to the conventions of the discipline: This grade will be awarded when the combined total marks from all assessment is at least 20% but less than 45% of total marks available.

Grade 3, Fail: Demonstrates superficial or partial or faulty understanding of the fundamental concepts of the field of study and limited ability to apply these concepts; presents undeveloped or inappropriate or unsupported arguments; communicates information or ideas with lack of clarity and inconsistent adherence to the conventions of the discipline: This grade will be awarded when the combined total marks from all assessment is at least 45% but less than 50% of total marks available.

Grade 4, Pass: Demonstrates adequate understanding and application of the fundamental concepts of the field of study; develops routine arguments or decisions and provides acceptable justification; communicates information and ideas adequately in terms of the conventions of the discipline: This grade will be awarded when the combined total marks from all assessment is at least 50% but less than 65% of total marks available.

Grade 5, Credit: Demonstrates substantial understanding of fundamental concepts of the field of study and ability to apply these concepts in a variety of contexts; develops or adapts convincing arguments and provides coherent justification; communicates information and ideas clearly and fluently in terms of the conventions of the discipline: This grade will be awarded when the combined total marks from all assessment is at least 65% but less than 75% of total marks available.

Grade 6, Distinction: As for 5, with frequent evidence of originality in defining and analysing issues or problems and in creating solutions; uses a level, style and means of communication appropriate to the discipline and the audience: This grade will be awarded when the combined total marks from all assessment is at least 75% but less than 85% of total marks available.

Grade 7, High Distinction: As for 6, with consistent evidence of substantial originality and insight in identifying, generating and communicating competing arguments, perspectives or problem solving approaches; critically evaluates problems, their solutions and implications: This grade will be awarded when the combined total marks from all assessment is greater than 85% of total marks available.

5.3 Late Submission

Late submissions will incur a penalty of 10% of the total marks received for each day that the submission is late (including weekend days). Submissions more than 5 days late will not be accepted. Late hardcopy submissions should be handed in to the lecturer via the ITEE School Office.

5.5 Assessment Detail

Assignment 1

Type: Problem Set/s
Learning Objectives Assessed: 1, 2, 3, 4
Due Date:
11 Apr 08
Weight: 20%
Task Description: Assignment 1
Criteria & Marking: This assignment will be based on questions from the practical sessions. Students will be awarded marks according to correctness of answers to practical questions. Presentation of answers (e.g code, figures, written responses) is also important when awarding marks.
Submission: Submission box.

Assignment 2

Type: Problem Set/s
Learning Objectives Assessed: 1, 2, 3, 4
Due Date:
16 May 08
Weight: 20%
Task Description: Assignment 2
Criteria & Marking: This assignment will be based on questions completed during the practical sessions. Students will be awarded marks according to correctness of answers to practical questions. Presentation of answers (e.g code, figures, written responses) is also important when awarding marks.
Submission: Submission box.

Final Exam

Type: Exam - during Exam Period (Central)
Learning Objectives Assessed: 3, 4, 5, 6, 7, 8
Due Date:
Examination Period
Weight: 60%
Perusal: 10 minutes
Duration: 120 minutes
Format: Short answer, Problem solving
Task Description: Final Examination
Criteria & Marking: Students will be awarded marks according to correctness of answers to closed-book questions. Part marks will be awarded for working.

6. Policies & Guidelines

This section contains the details of and links to the most relevant policies and course guidelines. For further details on University Policies please visit myAdvisor and the University Handbook of Policies and Procedures.

6.1 Assessment Related Policies and Guidelines

University Policies & Guidelines

An overview of the University’s assessment-related policies can be found on myAdvisor (http://www.uq.edu.au/myadvisor/index.html?page=2910).

Academic Integrity
It is the University's task to encourage ethical scholarship and to inform students and staff about the institutional standards of academic behaviour expected of them in learning, teaching and research. Students have a responsibility to maintain the highest standards of academic integrity in their work. Students must not cheat in examinations or other forms of assessment and must ensure they do not plagiarise.

Plagiarism
The University has adopted the following definition of plagiarism:

Plagiarism is the act of misrepresenting as one's own original work the ideas, interpretations, words or creative works of another. These include published and unpublished documents, designs, music, sounds, images, photographs, computer codes and ideas gained through working in a group. These ideas, interpretations, words or works may be found in print and/or electronic media.

Students are encouraged to read the UQ Academic Integrity and Plagiarism policy (http://www.uq.edu.au/hupp/index.html?page=25128) which makes a comprehensive statement about the University's approach to plagiarism, including the approved use of plagiarism detection software, the consequences of plagiarism and the principles associated with preventing plagiarism.

Feedback on Assessment
Feedback is essential to effective learning and students can expect to receive appropriate and timely feedback on all assessment. For a detailed explanation of the feedback you are entitled to, you should consult the policy on Student Access to Feedback on Assessment. (http://www.uq.edu.au/hupp/index.html?page=25114&pid=25075)

As a student you have a responsibility to incorporate feedback into your learning; make use of the assessment criteria that you are given; be aware of the rules, policies and other documents related to assessment; and provide teachers with feedback on their assessment practices.

There are certain steps you can take if you feel your result does not reflect your performance. Please refer to the myAdvisor web site. (http://www.uq.edu.au/myadvisor/index.html?page=2953&pid=2910)

School of Information Technology and Electrical Engineering Assessment Guidelines

Misconduct

Further to the statement on academic integrity and plagiarism above, students are required to read and understand the ITEE policy on Student Misconduct (http://www.itee.uq.edu.au/about_ITEE/policies/student-misconduct.html).

Late Arrival or Non-attendance at Examinations

The policy and procedure for late arrival or non-attendance at centrally controlled examinations is set out in the University's Examinations policy (HUPP 3.30.5), sections 8 and 10.2.

The way in which late arrival at a School-controlled examination is dealt with will be at the discretion of the course coordinator, who may be guided by the policy for centrally controlled exams.

In the case that a student requests a special exam for a School-controlled exam, the request will be considered and, if allowed, the timing shall be determined by the course coordinator, in consultation with the School's Chief Examiner where necessary, and in accordance with HUPP 3.30.5. Unless otherwise indicated in the Course Profile, applications must be made in writing to the Head of School no later than one week after the exam. Late applications will not be accepted.

Examination Feedback

In addition to the advice above, students wishing to view examination answer scripts and/or question papers should consult with the School office (Room 217, General Purpose South Building [78], St Lucia; Room 218, Building 1, Ipswich) regarding arrangements. The ITEE policy on exam script viewing is available at http://study.itee.uq.edu.au/current_students/exam_script_viewing.html.

Supplementary Assessment

If you fail this course you may be eligible for supplementary assessment - see the general award rules and/or your program rules for details. You should note that even though you may be eligible for supplementary assessment under these rules, in some circumstances there may be no practical assessment that can be offered to allow you to meet the minimum passing requirements. These circumstances may include failure based on:

group or team based assessment;
attendance or class participation requirements;
laboratory-based assessment, where laboratories can't practically be made available after classes have finished;
project or thesis-based assessment, where a significant period of time would be required to undertake supplementary assessment;
progressive assessment, where subsequent assessment items build on earlier assessment items; or
multiple assessment items, where it is impractical to offer multiple supplementary assessment items.

If the course coordinator determines that there is no practical supplementary assessment that can be offered to allow you to improve your grade, then you will not be offered supplementary assessment and your grade will remain unchanged.

6.2 Other Policies and Guidelines

University Policies and Guidelines

Placement Courses
Students on a placement course – also known as a work placement, internship, industry study, industry experience, clinical practice, clinical placement, practical work, practicum, fieldwork, teaching practice – should refer to the University policy, Placement Courses (http://www.uq.edu.au/hupp/index.html?page=25120&pid=25075) for detailed information.

Working with Children
Students whose studies include a professional/work placement, internship, clinical practice, teaching practice or other similar activity which involves them in regular contact with children should refer to the University policy, Working with Children Check - "blue card" (http://www.uq.edu.au/hupp/index.html?page=25004&pid=24963) to find out how to apply for a ‘blue card’.

Students with a Disability
Any student with a disability who may require alternative academic arrangements, including assessment, in the course/program is encouraged to seek advice at the commencement of the semester from a Disability Adviser at Student Support Services. Refer to the University policy, Students with a Disability (Disability Action Plan) (http://www.uq.edu.au/hupp/index.html?page=25122&pid=25075) and to the policy on Special Arrangements for Examinations for Students with a Disability (http://www.uq.edu.au/hupp/index.html?page=25111&pid=25075

Where an adjustment is made to an accredited program, it is the responsibility of the relevant Faculty to liaise with professional and registration bodies regarding the acceptability of the change/s.

Occupational Health and Safety
Undergraduate Students (http://www.uq.edu.au/hupp/index.html?page=25055&pid=25015) and Postgraduate Students (http://www.uq.edu.au/hupp/index.html?page=25057&pid=25015) should be familiar with the University policies on occupational health and safety in the laboratory.

Other School of Information Technology and Electrical Engineering Guidelines

Ethical Clearance

If your course involves assignment or project work involving human subjects or human-related materials, you must investigate the need for ethical clearance and obtain it when required. Information on ethical clearance can be found at http://www.uq.edu.au/research/orps/index.html?page=5064&pid=5256.

Learning Summary

Below is a table showing the relationship between the learning objectives for this course and the broader graduate attributes developed, the learning activities used to develop each objective and the assessment task used to assess each objective.

Learning Objectives