42894 Advanced Machine Learning
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 2025 is available in the Archives.
Credit points: 2 cp
Result type: Grade and marks
Requisite(s): 42820 Machine Learning Foundations
Description
The Advanced Machine Learning subject offers an introduction of some advanced machine learning data models, algorithms and theoretical results. It is designed for learners with some existing familiarity with machine learning who are looking for a deeper understanding (both theory and hands-on) of statistical learning theory and empirical risk minimization to improve their ML models and algorithms.
Subject learning objectives (SLOs)
Upon successful completion of this subject students should be able to:
| 1. | Design machine learning algorithms with practical implementation for professional contexts. (C.1) |
|---|---|
Course intended learning outcomes (CILOs)
This subject also contributes specifically to the development of the following Course Intended Learning Outcomes (CILOs):
- Design Oriented: FEIT graduates apply problem solving, design thinking and decision-making methodologies in new contexts or to novel problems, to explore, test, analyse and synthesise complex ideas, theories or concepts. (C.1)
Teaching and learning strategies
The study will be both theoretical and practical. As for the theory, the subject provides a systematic view of the whole lifecycle of a data model, from the design motivation to the dynamics in the learning process and the evaluation and how reliable the evaluation results are. On the practical side, learners will be led to translate mathematical notions into data structures and programs in a digital computer, which allows them to ‘open the hood’ of the data models and get hands-on experience to examine piece-by-piece how the models perform learning. Subject presentation includes weekly synchronous one-hour online workshops facilitated by an expert UTS academic(s) supporting self-study and online (LMS) learning activities. Regular formative quizzes throughout the semester will allow students to gauge their progress.
Content (topics)
- Build data models
- Stack perceptron to neural networks
- Deep neural networks (DNN) architecture, toolset and learning practice
- Generalised linear models, kernel methods
- Learn data models (Gradient-based algorithm and optimisation)
- Backpropagation - the training of neural networks
- Constrained optimisation practice
- Improve model reliability
- DNN structures to enable learning stability
- Regularisation techniques
- Why learned models can be trusted? (Theory of the risk of learning-based models)
- Bias and Variance, Training and Test Evaluation
Assessment
Assessment task 1: Algorithm Implementation and Report
| Intent: | Demonstrate ability to implement the learning framework and learning algorithm. It reinforces the hands-on skill in implementation and evaluation of machine learning. |
|---|---|
| Objective(s): | This assessment task addresses the following subject learning objectives (SLOs): 1 This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs): C.1 |
| Type: | Report |
| Groupwork: | Individual |
| Weight: | 100% |
| Length: | 2,000 words |
Minimum requirements
In order to pass the microcredential, a learner must achieve an overall mark of 50% or more.