FIT3181 · Deep learning

Overview

Deep learning (DL) has been fuelling Artificial Intelligence (AI) and the Fourth Industrial Revolution in recent years. The success of DL in many applications, including generative AI such as ChatGPT or DALL·E, has gained rocketed attention and becomes a highly demanded skill across industries and sectors. It is transforming innovations, powering new applications and impact our society in everyday activities. In this unit, you will learn the foundations of deep learning theory within a broader context of machine learning. At the same time, you will gain hands-on practical skills on how to apply DL to real-world applications across a range of AI cognitive tasks in computer vision such as image and object recognition, in natural language processing such as text classification using deep neural embeddings. Learning activities will focus on understand the fundamental concepts in DL such as neural networks (NN), convolutional NN, backpropagation and optimisation for deep learning, adversarial robustness, attention mechanism, transformer, important concepts in deep generative AI (VAE, GAN), in combination with laboratory sessions to gain hands-on experiences.

Areas of study: Data Science, Computer Science, Artificial Intelligence

Offerings

Assessment

The Handbook lists an examination for this unit.

This unit has threshold mark hurdles. You must attempt all assessments, achieve at least 40% in the midterm exam and an overall unit mark of 50% or more to pass the unit. If you do not achieve the threshold mark, you will receive a fail grade (NH) and a maximum mark of 45 for the unit.

Assessment details may change. Please refer to the assessment information in Moodle closer to the start of the teaching period.

Requisites

Learning outcomes

1. Describe basic and advanced concepts of machine learning, AI, and deep learning
2. Assess what deep learning is, what makes deep learning work or fail, and critique where they should be applied.
3. Explain fundamental elements of deep learning.
4. Construct deep neural networks, convolutional NNs, RNN, deep generative models and apply different strategies for training them
5. Apply DL models in real-world applications such as image classification, text translation, image/text generation
6. Develop critical thinking and obtain hands-on experiences with practical deep learning models and frameworks

Workload

Minimum total expected workload to achieve the learning outcomes for this unit is 144 hours per semester typically comprising a mixture of scheduled online and face to face learning activities and independent study. Independent study may include associated reading and preparation for scheduled teaching activities.

Student experience, not official rules. Nothing here changes what the Handbook says.