| CODE | CCE5106 | ||||||||||||||||
| TITLE | Advanced Neural Network Models | ||||||||||||||||
| UM LEVEL | 05 - Postgraduate Modular Diploma or Degree Course | ||||||||||||||||
| MQF LEVEL | 7 | ||||||||||||||||
| ECTS CREDITS | 5 | ||||||||||||||||
| DEPARTMENT | Communications and Computer Engineering | ||||||||||||||||
| DESCRIPTION | This study-unit covers theoretical and practical aspects of advanced pattern recognition and machine learning techniques, namely deep learning. Both the model itself and the learning methods appropriate for the various architectures studied are covered, including supervised, reinforcement learning and self-supervised learning. Topics covered in this unit fall under the families of: recurrent neural networks (RNN) and convolutional neural networks (CNN), and include, autoencoders (AEs), generative adversarial networks (GANs), reinforcement learning in deep learning (RL-DL), gated recurrent unit (GRUs), long short term memory models (LSTMs) models with attention mechanisms, transformers, bi-directional models (Bi-LSTM, BERT) and joint models. Examples of application areas where the models are applied, mainly in vision and language, but not only, are discussed throughout. Study-unit Aims: - To provide a theoretical foundation and framework to the practice of defining the architecture of deep neural networks; - To provide a theoretical foundation on the various techniques and methods used during the process of training deep neural networks; - To discuss example models as applied to vision and language tasks; - To provide students with an opportunity to enhance the theoretical knowledge gained via programming assignments. Learning Outcomes: 1. Knowledge & Understanding: By the end of the study-unit the student will be able to: - Describe basic and advanced deep learning architectures and techniques; - Describe a number of learning or optimisation techniques for training the models; - Describe some applications of deep learning models. 2. Skills: By the end of the study-unit the student will be able to: - Select, Develop and/or design deep learning architectures to solve problems; - Select appropriate loss functions and optimisers; - Train and debug deep learning architectures; - Evaluate the deep learning model. Main Text/s and any supplementary readings: Main Text: I. Goodfellow, Y. Bengio and A. Courville, “Deep Learning”, MIT, 2016. |
||||||||||||||||
| ADDITIONAL NOTES | This study-unit builds on a first course in machine learning, e.g CCE5107/8, CCE5225 or equivalent. | ||||||||||||||||
| STUDY-UNIT TYPE | Lecture | ||||||||||||||||
| METHOD OF ASSESSMENT |
|
||||||||||||||||
| LECTURER/S | Adrian F. Muscat Trevor Spiteri |
||||||||||||||||
|
The University makes every effort to ensure that the published Courses Plans, Programmes of Study and Study-Unit information are complete and up-to-date at the time of publication. The University reserves the right to make changes in case errors are detected after publication.
The availability of optional units may be subject to timetabling constraints. Units not attracting a sufficient number of registrations may be withdrawn without notice. It should be noted that all the information in the description above applies to study-units available during the academic year 2023/4. It may be subject to change in subsequent years. |
|||||||||||||||||