49275 Neural Networks and Fuzzy Logic
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.
Credit points: 6 cp
Subject level:
Postgraduate
Result type: Grade and marksRequisite(s): 120 credit points of completed study in Bachelor's Honours Embedded Degree owned by FEIT OR 120 credit points of completed study in Bachelor's Combined Honours Degree owned by FEIT OR 120 credit points of completed study in Bachelor's Combined Honours Degree co-owned by FEIT
These requisites may not apply to students in certain courses. See access conditions.
Recommended studies:
introductory programming or introductory control subjects
Description
The principal objective of this subject is to introduce students to neural networks and fuzzy theory from an engineering perspective. This is a hands-on subject where students are given integrated exposure to professional practice. These areas include identification and control of dynamic systems, neural networks and fuzzy systems can be implemented as model-free estimators and/or controllers. As trainable dynamic systems, these intelligent control systems can learn from experience with numerical and linguistic sample data. As an example, students will develop an expertise in biomedical, pattern recognition, control system using neural networks and fuzzy logic.
Subject learning objectives (SLOs)
Upon successful completion of this subject students should be able to:
| 1. | Gain proficiency in fundamental and advanced theories related to neural networks and fuzzy logic, and apply this knowledge to address real-world engineering problems. (D.1) |
|---|---|
| 2. | Apply analytical thinking, research methodologies, and advanced problem-solving skills to successfully implement a neural network, convolutional deep network, or fuzzy logic system, addressing complex real-world engineering challenges. (C.1) |
| 3. | Collaborate to manage and conduct research projects using neural networks and fuzzy logic. (E.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)
- Technically Proficient: FEIT graduates apply theoretical, conceptual, software and physical tools and advanced discipline knowledge to research, evaluate and predict future performance of systems characterised by complexity. (D.1)
- Collaborative and Communicative: FEIT graduates work as an effective member or leader of diverse teams, communicating effectively and operating autonomously within cross-disciplinary and cross-cultural contexts in the workplace. (E.1)
Contribution to the development of graduate attributes
Engineers Australia Stage 1 Competencies
Students enrolled in the Master of Professional Engineering should note that this subject contributes to the development of the following Engineers Australia Stage 1 competencies:
- 1.3. In-depth understanding of specialist bodies of knowledge within the engineering discipline.
- 1.4. Discernment of knowledge development and research directions within the engineering discipline.
- 2.1. Application of established engineering methods to complex engineering problem solving.
- 2.2. Fluent application of engineering techniques, tools and resources.
- 3.3. Creative, innovative and pro-active demeanour.
- 3.6. Effective team membership and team leadership.
Teaching and learning strategies
There are several modules in this subject: background to neural networks and fuzzy logic, intelligent systems and advanced applications.
As this is a hands-on real-world practice oriented subject, active learning opportunities are integrated into the class. These include working on a real-world problems in small collaborative teams, to identify a problem and then ideate a solution. Each week, team members will be mentored by either academics or tutors to explore new idea and develop solutions in the context of their neural networks.
The subject is highly collaborative, and teams will be formed in week 1 with a maximum of 3 students in a team. All teams will select a joint project based on their interest and will conduct research and work together for the duration of the subject. Students will then deliver their project solution in a verbal seminar and research papers, explaining their interpretations and explanations.
These projects may include the development of a computer software system (e.g. a handwritten character recognition system, or traffic sign classification), a control system (e.g. a ball and beam system) or a biomedical system (e.g. wheelchair control, medical imaging, brain computer interface system).
Lectures, tutorials and laboratories are used to discuss this information, to answer any questions and to provide support and ongoing feedback.
Content (topics)
The first module will discuss the fundamental concepts of artificial neural systems and fuzzy logic theory. It introduces the foundations of neural network learning principles including the Hebbian, Perceptron, Delta, and Widrow-Hoff learning rules. It includes fuzzy sets, linguistic variables and approximate reasoning.
The second module covers the concept of error function 3eneralized3 using the steepest descent 3eneralized3 technique. This will then be extended to multilayer feedforward neural networks. Other concepts such as 3eneralized delta learning rule, feedforward recall and error back-propagation training, associative memories, matching and self-organising networks will be discussed. Further topics will include the preliminaries and basic construction of a fuzzy controller, adaptive fuzzy logic controller, and the self-organising fuzzy logic controller. Lastly, a basic concept of convolutional neural network and genetic algorithm are introduced.
The third module relates to advanced applications of intelligent systems. The design and development of a semi-autonomous wheelchair using neural network classifications will be discussed. Other advanced applications relate to the control of blood glucose in patients with diabetes using a fuzzy system.
Assessment
Assessment task 1: Assignment 1: Basic Neural Networks and Fuzzy Systems
| Intent: | To analyse basic concepts within neural networks and fuzzy logic by solving real-world problems. |
|---|---|
| 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): D.1 |
| Type: | Report |
| Groupwork: | Individual |
| Weight: | 20% |
| Length: | Approximately 700-1000 words, comprising explanations, discussions, and step-by-step procedures (excluding source code, figures, and tables from the word count). |
Assessment task 2: Assignment 2: Advanced Neural Networks and Fuzzy Systems
| Intent: | To apply advanced theories of neural networks and fuzzy logic by solving real-world problems. |
|---|---|
| Objective(s): | This assessment task addresses the following subject learning objectives (SLOs): 2 This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs): C.1 |
| Type: | Report |
| Groupwork: | Individual |
| Weight: | 20% |
| Length: | Approximately 700-1000 words, comprising explanations, discussions, and step-by-step procedures (excluding source code, figures, and tables from the word count). |
Assessment task 3: Reflective Essays
| Intent: | Aims to reflect and apply the knowledge you have learned in the seminars by lecturer/guest speakers. |
|---|---|
| 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): D.1 |
| Type: | Essay |
| Groupwork: | Individual |
| Weight: | 10% |
| Length: | Approximately 500 words for each submission |
Assessment task 4: Major Group Project
| Intent: | Develop and implement a basic trainable neural network or a fuzzy logic system for a typical control, computing application or biomedical application. |
|---|---|
| Objective(s): | This assessment task addresses the following subject learning objectives (SLOs): 1, 2 and 3 This assessment task contributes to the development of the following Course Intended Learning Outcomes (CILOs): C.1, D.1 and E.1 |
| Type: | Project |
| Groupwork: | Group, group and individually assessed |
| Weight: | 50% |
| Length: | Research article: 2000-2500 words, double column (IEEE format) |
| Criteria: | 50% (Seminar 1: 10%; Seminar 2/Demo: 10%; Research Paper: 30%) |
Minimum requirements
In order to pass the subject, a student must achieve an overall mark of 50% or more.
Required texts
Ling S, "Neural Networks and Fuzzy Logic", UTS, 2019
Recommended texts
Zurada J.M., Introduction to Artificial Neural Systems, West Publishing Company, 1992
(alternative) Bishop C., Neural Networks for Pattern Recognition, Oxford Univ. Press, 2004
Ross T.J., Fuzzy Logic with Engineering Applications, John Wiley & Sons, 2010
Michalewicz Z, "Genetic Algorithm+Data Structures=Evolution Programs" Spring-Verlag Berlin, 1996
Goodfellow et. Al, Deep Learning, MIT Press Academic, 2014
References
Öztürk S., Convolutional Neural Networks for Medical Image Processing Applications, Taylor & Francis Ltd, 2022
Ayman S. El-Baz, Suri J., State of the art in neural networks and their applications, Academic Press, 2021.
Cooper S., Neural Networks- A Practical Guide For Understanding And Programming Neural Networks And Useful Insights For Inspiring Reinvention, IGHTNING SOURCE INC, 2019.Haykin S., Neural Networks - A Comprehensive Foundation, Prentice Hall, 1999
Katagiri S., Handbook of Neural Networks for Speech Processing, Artech House, 2000
Ling SH, "Genetic Algorithm and Variable Neural Networks: Theory and Application", Lambert Academic Publishing, 2010
Jang J.S.R., Sun C.T., Mizutani E., Neuro-Fuzzy and Soft Computing, Prentice Hall, 1997
Yan J., Ryan M., Power J., Using Fuzzy Logic, Prentice Hall, 1994
Brown M., Neurofuzzy Adaptive Modelling and Control, Prentice Hall, 1994
Wang L.X., Adaptive Fuzzy Systems and Control, Prentice Hall, 1994
Yager R.R., Essential of Fuzzy Modelling and Control, John Wiley, 1994
Kosko B., Neural Networks and Fuzzy Systems, Prentice-Hall, 1992
McNeill D. and Freiberger P., Fuzzy Logic, Bookman Press, 1993
Other resources
Canvas
All students will have an account on Canvas. They are expected to check the site regularly for announcements and information. Additional resources are given in Canvas for specific major projects. These consist of a number of journal and conference publications relating these major projects.