University of Technology Sydney

37401 Machine Learning: Mathematical Theory and Applications

Warning: The information on this page is indicative. The subject outline for a particular session, location and mode of offering is the authoritative source of all information about the subject for that offering. Required texts, recommended texts and references in particular are likely to change. Students will be provided with a subject outline once they enrol in the subject.

Subject handbook information prior to 2024 is available in the Archives.

UTS: Science: Mathematical and Physical Sciences
Credit points: 8 cp
Result type: Grade and marks

There are course requisites for this subject. See access conditions.

Description

Machine learning is a field of artificial intelligence that enables machines to learn from data. This subject is designed to provide a comprehensive introduction to the fundamentals of machine learning and their applications in engineering, statistics, computer science, finance, and economics. Students gain an in-depth understanding of the techniques and applications of machine learning. In particular, the subject covers modern topics such as supervised learning, semi-supervised learning, unsupervised learning, regularisation, decision and regression trees, ensemble methods, Gaussian processes, and deep learning. The subject is designed to provide students with hands-on experience in implementing and using machine learning algorithms, as well as popular packages and libraries.

Subject learning objectives (SLOs)

Upon successful completion of this subject students should be able to:

1. Explain the trade-off between bias and variance for a range of machine learning models
2. Evaluate a machine learning model based on its predictive performance
3. Formulate an appropriate model for a given dataset and determine an estimation approach
4. Use programming to implement an algorithm to estimate a model and evaluate its performance.
5. Compare and contrast solutions relevant to real world applications of machine learning in written format.
6. Demonstrate collaborative skills to solve problems in machine learning

Course intended learning outcomes (CILOs)

This subject also contributes specifically to the development of following course intended learning outcomes:

  • Demonstrate critical appraisal of advanced knowledge and critically evaluate the informationís source and relevance, with a focus on applications of mathematical and statistical methodologies to problem solving. (1.1)
  • Tackle the challenge of complex real-world problems in the areas of mathematical and statistical modelling by critically evaluating information and solutions and conducting appropriate approaches to independent research. (2.1)
  • Engage in work practices that demonstrate an understanding of confidentiality requirements, ethical conducts, data management, and organisation and collaborative skills in the context of applying mathematical and statistical modelling. (3.1)
  • Find and reflect on the value, integrity, and relevance of multiple sources of information to derive innovative solutions, show creativity, innovation and application of technologies in evaluating solutions to contemporary mathematics problems. (4.1)
  • Identify and present complex ideas and justifications using appropriate communication approaches from a variety of methods (oral, written, visual) to communicate with mathematicians, data analysts, scientists, industry, and the general public. (5.1)

Contribution to the development of graduate attributes

The subject contributes to the following course intended learning outcomes.

Graduate attribute 1.0: Disciplinary knowledge

The models, theorems, and implementations presented in the subject provide students with disciplinary knowledge of machine learning and its application to problems in science. Students will learn state-of-the-art methods that are widely used in practice, thereby empowering them with professional skills that are sought after in the industry, the public sector, and academia.

Graduate attribute 2.0: Research, inquiry and critical thinking

Machine learning methods at the research frontier will be studied and critically assessed. Students will study the bias-variance tradeoff underlying complex models, which provides an understanding of how to design and implement models suitable for real-world problems, and how these can be critically evaluated.

Graduate attribute 3.0: Professional, ethical, and social responsibility

The subject involves multiple tasks that involve teamwork and trains the students to work effectively and efficiently in a team. The subject teaches the students the importance of reproducible research, which is facilitated by implementing methods in a notebook environment with a structured workflow.

Graduate attribute 4.0: Reflection, Innovation, Creativity

The assessment tasks in this subject are designed to train the students' ability for creative problem-solving. The students will be trained in applying and creatively combining methods from the statistical and machine learning literature to solve industry-relevant problems.

Graduate attribute 5.0: Communication

The solutions to the assessment tasks are presented in report form that documents the code as well as the main findings. The subject thus trains the students to effectively communicate the analysis of the output from advanced machine learning methods.

Teaching and learning strategies

The subject is taught through a combination of interactive sessions, computer labs solving collaborative assignments to be handed in, and a final individual project solved under a shorter time constraint.

During the interactive sessions, the lecturer will present and discuss the theory. Students are encouraged to contribute to the discussions and ask for any clarifications needed. Recorded more detailed material will be used to enhance the learning experience.

The computer labs and hand-in assignments will test the students' understanding of the theory and their ability to apply the theory to solve industry relevant problems collaborating in small groups. The problems will be solved by a mix of the students implementing the algorithms and them applying software packages, where the former aims to facilitate the understanding of the methods, while the latter provides the students with state-of-the art tools to solve real world problems.

The final project will test the students ability to individually apply the learned skills to solve real world problems in a time constrained manner.

Content (topics)

The subject presents a number of popular methods in machine learning and a general approach to building and evaluating predictive models. The underlying ideas and mathematical intuition of the methods are emphasised, as well as their implementation. The following topics will be covered:

- Supervised learning via distance-based (e.g. k-nearest neighbours) and rule-based methods (e.g. decision trees)

- Regularised estimation methods for regression and classification.

- The bias-variance trade-off.

- Learning parametric models.

- Neural networks and deep learning.

- Ensemble methods.

- Gaussian processes for regression and classification.

- Generative modelling and unsupervised learning.

Assessment

Assessment task 1: Computer labs (group)

Intent:

This assessment task contributes to the development of the following graduate attributes:

1. Disciplinary Knowledge

2. Research, inquiry, and critical thinking

3. Professional, ethical and social responsibility

4. Reflection, Innovation, Creativity

5. Communication

Objective(s):

This assessment task addresses subject learning objective(s):

1, 2, 3, 4, 5 and 6

This assessment task contributes to the development of course intended learning outcome(s):

1.1, 2.1, 3.1, 4.1 and 5.1

Type: Report
Groupwork: Group, group assessed
Weight: 30%
Criteria:

Correct answers to questions about the module topics.

Assessment task 2: Computer labs (individual)

Intent:

This assessment task contributes to the development of the following graduate attributes:

1. Disciplinary Knowledge

2. Research, inquiry, and critical thinking

4. Reflection, Innovation, Creativity

5. Communication

Objective(s):

This assessment task addresses subject learning objective(s):

1, 2, 3, 4 and 5

This assessment task contributes to the development of course intended learning outcome(s):

1.1, 2.1, 4.1 and 5.1

Type: Report
Groupwork: Group, individually assessed
Weight: 30%
Criteria:

Correct answers to questions about the module topics.

Assessment task 3: Final project

Intent:

This assessment task contributes to the development of the following graduate attributes:

1. Disciplinary Knowledge

2. Research, inquiry, and critical thinking

4. Reflection, Innovation, Creativity

Objective(s):

This assessment task addresses subject learning objective(s):

1, 2, 3 and 4

This assessment task contributes to the development of course intended learning outcome(s):

1.1, 2.1 and 4.1

Type: Quiz/test
Groupwork: Individual
Weight: 40%
Criteria:

Correct answers to questions about the module topics.

Minimum requirements

Attendance is critical to succeed in the final project which constitutes 40% of the total mark. Students must obtain an overall mark of at least 50 to pass this subject.