Use the Tab and Up, Down arrow keys to select menu items.
This course introduces students to the core concepts of deep neural networks. The course focuses on the computational process of problem formulation, model selection and design, implementation, analysis, and refinement for deep neural networks. We analyze a range of advanced neural network designs with transformative results in computer vision, natural language, anomaly detection, molecular design, and deep fakes. Students build competency in the theory and practice of creating deep neural network applications and will research, propose, and implement their own deep learning network for a given application domain.
1. Analyse the fundamental algorithms of deep learning systems such as backpropagation [WA1, WA2, WA10]2. Select and analyse advanced neural network architectures for processing and generating images, video, text, audio, time sequences, and graphs [WA2, WA6, WA10]3. Constructing and implement deep learning models through programming [WA3, WA5]4. Train neural networks through supervised learning, unsupervised learning, and reinforcement learning [WA3, WA4, WA5]5. Refine and engineer models for domain constraints such as performance in edge systems [WA3, WA4, WA5, WA8]
(1) 30 points of 300-level COSC/SENG/DATA/ENCE/ENEL301; and (2) COSC122; and (3) COSC262 or ENEL300 or ENMT301
Students must attend one activity from each section.
Please note that the course activity times advertised here are currently in draft form, to be finalised at the end of January for S1 and whole year courses, and at the end of June for S2 courses.Please hold off enquiries about these times until those finalisation dates.
James Atlas
The Computer Science department's grading policy states that in order to pass a course you must meet two requirements:1. You must achieve an average grade of at least 50% over all assessment items.2. You must achieve an average mark of at least 45% on invigilated assessment items.If you satisfy both these criteria, your grade will be determined by the following University-wide scale for converting marks to grades: an average mark of 50% is sufficient for a C- grade, an average mark of 55% earns a C grade, 60% earns a C+ grade and so forth. However if you do not satisfy both the passing criteria you will be given either a D or E grade depending on marks. Marks are sometimes scaled to achieve consistency between courses from year to year.Students may apply for special consideration if their performance in an assessment is affected by extenuating circumstances beyond their control.Applications for special consideration should be submitted via the Special Considerations website within five days of the assessment.Where an extension may be granted for an assessment, this will be decided by direct application to the Department and an application to the Examinations Office may not be required. Special consideration is not available for items worth less than 10% of the course.Students prevented by extenuating circumstances from completing the course after the final date for withdrawing, may apply for special consideration for late discontinuation of the course. Applications must be submitted to the Examinations Office within five days of the end of the main examination period for the semester.
• Machine learning concepts• Types of learning (supervised, self-supervised, reinforcement) • Types of problems (classification, regression)• Loss functions, gradient descent and optimization• Automatic differentiation, forward and backward mode• Diagnosing problems: under and over-fitting, regularization, initialization • Deep learning concepts• Multi-dimensional arrays and memory models, views and vectorized operations• Neural networks: perceptron, layers, types of operations (linear, convolutional, pooling, sampling, nonlinearity), visualization• Sequential and recurrent networks• Transfer learning, synthesis, ensemble networks• Deep learning problems, models, and research• Computer Graphics and Vision • object detection, segmentation, image retrieval, face reidentification• feature pyramid networks, similarity learning, discriminatory networks, adversarial networks, generative networks• Natural language• parsing, window prediction, generation, translation• encoders/decoders, latent space, autoencoders • transformers, long short-term memory, attention• Audio and Video Synthesis• Text-to-speech, music generation, deep fakes, semantic models• Time series forecasting, autoregression, dilated convolution, few-shot learning• Search using deep reinforcement learning• Molecular design, game playing• neural network architecture design, compression, quantization• Agents, Markov decision processes, Monte Carlo, policy gradient methods• Anomaly detection• Intrusion detection, fraud detection, scientific discovery• One-class neural networks, zero shot learning• Irregular networks• Recommender systems, molecular structure and property prediction• Graph convolutional networks, point cloud processing, spatial-temporal networks
Please click HERE for the CSSE Department's policy for the academic remedy of applications for a special consideration for final exams.
Domestic fee $1,176.00
International Postgraduate fees
* All fees are inclusive of NZ GST or any equivalent overseas tax, and do not include any programme level discount or additional course-related expenses.
For further information see Computer Science and Software Engineering .