Honours and Minor Thesis projects

Multi-label classification (MLC), which simultaneously assigns several labels to each instance, is critical in a wide variety of domains. One of the most difficult is a subset of data categorisation in which classes are arranged hierarchically and objects can be allocated to many paths of the class hierarchy concurrently. This is referred to as hierarchical multi-label classification (HMC), and it is useful for text classification. For example, the output of a news article may cover a variety of topics, including news, finance, and sports.

In this project, you will build an autonomous agent in the MineRL environment for playing Minecraft or an agent for Animal-AI.  Herein, you will learn how to incorporate symbolic prior knowledge for improving the performance of an agent trained by using deep reinforcement learning (RL) technique, which is the core technique to build AlphaGo.

For the numerous benefits online shopping provides, their popularity has been on rise. Moreover, due to the recent COVID-19 pandemic online shopping has become a regular activity in most of the households. As such e-commerce application are being developed at different scale. All those may not understand the requirement properly and sometimes may have overlooked special end user needs. This work is designed to understand the requirements of the e-commerce applications and how personal characteristics of the end users can influence the requirements.…

It is quite challenging to access to videos for people who are blind or have low vision (BLV), particularly creating audio descriptions that describe the scenes without interfering the dialogues in a video. There is also the challenge of providing additional information using multi-modal feedback, that is using non-speech audio and haptics.

People who are blind or have low vision (BLV) access documents using screen readers such as JAWS and NVDA. These screen readers emulates a cursor moving around the screen using arrow keys or various shortcut combinations.

In this project you will work on creating a 3D printed platform used with an iPad for people who are blind or have low vision. The platform will allow people to program in the Scratch visual programming language (https://scratch.mit.edu/) using 3D printed blocks.

With the rise of software systems ranging from personal assistance to the nation's facilities, software defects become more critical concerns as they can cost millions of dollars as well as impact human lives. Yet, at the breakneck pace of rapid software development settings (like CI/CD, Agile, Rapid Releases), Software Quality Assurance (QA) practices (e.g., code review and software testing) nowadays are still time-consuming.

Text classification has extensive uses and deep learning has improved its performance using transformer language models.  A major hurdle for its use, however, is the paucity of labelled or annotated data. The data labelling process performed by domain experts is expensive and tedious to produce.  Active Learning is an approach to speeding up learning by judiciously selecting data to be annotated. Recently, advances in active learning theory have been made, but some experimental anomalies occur which need investigating.

In this project, the goal is to develop a new method (using computer vision and machine learning techniques) for robotic navigation in which goals can be specified at a much higher level of abstraction than has previously been possible. This will be achieved using deep learning to make informed predictions about a scene layout and navigating as an active observer in which the predictions inform actions.

Admission to the Medical degree at Monash is very competitive. The criteria is beyond the ATAR score. It covers many other factors, such as health test, interview, etc. Consequently, the ranking process is very complex.