COMP328 Fundamentals of Machine Learning

Open Online Course

COMP328 Fundamentals of Machine Learning

A full-semester introduction to machine learning — from data preparation and classical algorithms through neural networks, deployment, and responsible AI. Built around a complete, hands-on learning path in Python.

Start the course Explore the curriculum
3 credits 6 modules 20 units Online, asynchronous

Overview

Learn to build, evaluate, and question intelligent systems

COMP328 treats machine learning as both a technical discipline and a professional responsibility. You will build models that work — and learn to recognise when they should not be trusted.

This course provides a comprehensive introduction to the principles, algorithms, and practices of machine learning. You begin with the conceptual and mathematical foundations of learning from data, then move systematically through supervised learning, classical algorithms, unsupervised methods, and neural networks.

Every unit pairs explanation with implementation. s include worked examples, visual illustrations, interactive activities, and self-check questions, so you can move at your own pace while still building genuine practical fluency.

The course closes with deployment, ethics, and governance — because a model that never ships, or ships unfairly, has not solved a real problem.

Knowledge

  • Explain core ML principles and differentiate learning paradigms.
  • Describe the foundations of linear models, tree-based methods, and neural networks.

Skills

  • Prepare and explore real datasets with NumPy, Pandas, and Scikit-Learn.
  • Implement, train, and evaluate regression, classification, clustering, and dimensionality reduction models.
  • Interpret results using appropriate metrics and diagnose bias and variance.
  • Design and execute a complete ML project from definition to presentation.

Professional Competencies

  • Assess ethical implications, bias, and fairness concerns.
  • Communicate ML findings to technical and non-technical audiences.

Curriculum

Six modules, twenty units, one coherent path

Each module builds on the last. Start at Module 1 and work forward — or jump directly to a unit using the links below.

How it works

Every follows the same nine-part structure

Consistency is deliberate. Once you know the rhythm of a lesson, you can focus entirely on the ideas rather than on navigating the material.

Anatomy of a

Each of the 170+ s is self-contained and follows an identical instructional design.

  1. Learning ObjectivesWhat you should be able to do by the end.
  2. Motivating ExampleA concrete problem that makes the concept matter.
  3. Concept ExplanationThe core idea, explained clearly and precisely.
  4. Visual IllustrationDiagrams and flowcharts for intuition.
  5. Worked ExampleA step-by-step walkthrough you can follow.
  6. Interactive ActivityPractice with code or guided exploration.
  7. Self-Check QuestionsVerify your understanding before moving on.
  8. SummaryThe key takeaways, condensed.
  9. Further ReadingWhere to go deeper, if you want to.

Assessment

How you will be graded

Assessment is weighted toward applied work. The capstone and assignments together account for 80% of your final grade.

15%

Unit Quizzes

Twenty end-of-unit quizzes covering the core concepts from each unit’s readings and activities.

40%

Assignments

Five hands-on projects applying ML techniques to real-world datasets, from exploration to neural networks.

40%

Capstone Project

A complete ML lifecycle project: problem selection through to final report and presentation.

5%

Participation

Constructive engagement in discussion forums and peer review activities.

Assignments

  • Assignment 1 Data Exploration and Preparation Open
  • Assignment 2 Regression Modeling and Evaluation Open
  • Assignment 3 Classification Analysis Open
  • Assignment 4 Unsupervised Learning Techniques Open
  • Assignment 5 Neural Network Modeling Open
  • Capstone End-to-end machine learning project Open

Tools & environment

Everything runs on a standard Python stack

No specialised hardware is required. All practical work can be completed on a modern laptop using free, open-source software.

Python 3.9+ The language for all course code and examples.
JupyterLab Notebook environment for lessons and assignments.
NumPy & Pandas Numerical Computing and data manipulation.
Matplotlib Visualisation for exploration and results.
Scikit-Learn Classical machine learning algorithms.
TensorFlow / Keras Neural network and deep learning units.

Getting started

Four steps to begin

Set aside about 30 minutes to get oriented before diving into Unit 1.

  1. Read the orientation

    Start with the Welcome & Orientation page for course structure, expectations, and study advice.

  2. Review the syllabus

    Check the syllabus for learning outcomes, assessment weightings, policies, and the full schedule.

  3. Set up your environment

    Install Python, JupyterLab, and the core libraries. Unit 3 covers this in detail if you would rather learn it in context.

  4. Begin Module 1

    Open Module 1: Foundations of Machine Learning and work through Unit 1 at your own pace.

Questions

Frequently asked

What background do I need before starting?

You should be comfortable with introductory programming, basic Python, introductory statistics, and basic algebra. Equivalent prerequisites are COMP 200 Programming Fundamentals and COMP 268 Introduction to Statistics.

If your Python is a little rusty, Unit 3 (Python ML Ecosystem) is designed to bring you up to speed before the modelling begins.

How much time should I expect to spend each week?

COMP328 is a 3-credit course, so plan for roughly 8–10 hours per week across reading, activities, quizzes, and assignments. Weeks with an assignment due will be heavier.

Do I need a GPU or a powerful computer?

No. All coursework is designed to run on a standard modern laptop using free, open-source software. The neural network units use small datasets and modest architectures that train comfortably on CPU.

Is there a required textbook?

There is no single required textbook. All necessary readings, articles, and documentation are provided within the course modules, with further-reading suggestions at the end of each .

How is the course assessed?
  • Unit quizzes (20) — 15%
  • Assignments (5) — 40%
  • Capstone project — 40%
  • Participation — 5%

See the syllabus for full grading details and the letter-grade scale.

Can I work through the modules out of order?

The modules are designed to build on one another, and later units assume the vocabulary and techniques introduced earlier. That said, each is self-contained, so if you already have a strong background you can navigate directly to the units you need.

What will I have to show at the end of the course?

The capstone project integrates the complete machine learning lifecycle: problem selection, data acquisition, data cleaning, model building, model evaluation, interpretation, ethics analysis, a final written report, and a presentation.

Ready to begin?

Start with the orientation page, then open Module 1. The whole path is laid out for you — one unit at a time.

Orientation Module 1
👤