Introduction to Deep Learning

Course aim

To develop knowledge of the fundamental principles, theory and approaches for learning with deep neural networks.

Course content

This course will teach students the fundamentals of modern multi-layered neural networks. It will cover the most common forms of model architectures and primarily the algorithms used to train them. The theory and principles will be presented, but this will go hand-in-hand with a focus on practical experience such as using existing frameworks and implementing (simplified versions) of core algorithms.

Students will be taught the basics of neural networks, convolutional networks, recurrent networks; and introduced to concepts such as: dropout, batch normalization, types of hyper-parameter optimization, distributed and constrained computing variants. Applications in the area of audio processing and image captioning and vision will be discussed.

Textbook(s)

Nil

Prerequisite(s)

Nil

Corequisite(s)

Nil

Teaching method

Note: These components may or may not be scheduled in every study period. Please refer to the timetable for further details.

Assessment

Assignment

Fees

EFTSL*: 0.125
Commonwealth Supported program (Band 2)
To determine the fee for this course as part of a Commonwealth Supported program, go to:
How to determine your Commonwealth Supported course fee. (Opens new window)

Fee-paying program for domestic and international students
International students and students undertaking this course as part of a postgraduate fee paying program must refer to the relevant program home page to determine the cost for undertaking this course.

Non-award enrolment
Non-award tuition fees are set by the university. To determine the cost of this course, go to:
How to determine the relevant non award tuition fee. (Opens new window)

Not all courses are available on all of the above bases, and students must check to ensure that they are permitted to enrol in a particular course.

* Equivalent Full Time Study Load. Please note: all EFTSL values are published and calculated at ten decimal places. Values are displayed to three decimal places for ease of interpretation.