This project takes a different approach to RL, inspired by evidence that Hippocampus replays to the frontal cortex directly. It is likely used for model building, as opposed to the mainstream view in cognitive science and ML - where 'experience replay' ultimately improves policy. The predicted benefits are sample efficiency, better ability to generalize to new tasks and an ability to learn new tasks without forgetting old ones.
Honours and Minor Thesis projects
The brains of all bilaterally symmetric animals on Earth are divided into left and right hemispheres. The anatomy and functionality of the hemispheres have a large degree of overlap, but they specialize to possess different attributes. This principle is poorly understood and has not been exploited in AI/ML. The right hemisphere is more dominant for novelty, and the left for routine. Activity slowly moves to the left hemisphere as a task is perfected. In this project, we apply that principle to continual RL, where new tasks are introduced over time.
The hippocampus is critical for episodic memory, a key component of intelligence, and a sense of self. There are a number of computational models, but none of them consider the fact that the hippocampus is, like the rest of the brain, divided into Left and Right hemispheres. Division into Left and Right is poorly understood, but undoubtedly critical, as it is a remarkably conserved feature of all bilaterally symmetric animals on Earth.
The brains of all bilaterally symmetric animals on Earth are divided into left and right hemispheres. The anatomy and functionality of the hemispheres have a large degree of overlap, but they specialize to possess different attributes. This principle is poorly understood and has not been exploited in AI/ML. Previously, we mimicked biological differences between hemispheres, and achieved specialization and superior performance in a classification task that matched behavioral observations.
Are you interested in applying your database knowledge to a real project? This project aims to develop a patient registry for hospitals around Australia. This is a collaboration with the Faculty of Medicine, Monash University. We will be building a central database or a data warehouse repository to store patient admission to the hospitals.
Smart TV has become the dominant TV type nowadays. More and more users are switching from traditional TVs to Smart TVs. Despite the growing momentum of the smart TV industry (particularly in terms of the number of TV devices accessible in the Android ecosystem), the number of currently available TV apps is significantly less than the number of existing smartphone apps. There is an easily overlooked gap between the smartphone developers and smart TV (hereafter, TV) apps, leaving the prospect of TV apps behind the smartphone.
Rheumatoid arthritis (RA) is a chronic, inflammatory disorder that if untreated will lead to irreversible destruction of the joints.
Poor food choices could highly contribute to the development of the current epidemic of diabetes. The consumption of foods that result in a large increase in blood glucose after consumption does lead to a progressive decline in beta cell function as well as contributing to obesity and insulin resistance leading to type 2 diabetes. Mobile apps could provide useful features for automatically identify, recording and monitoring of food intake, using food item detection methods.
This project aims to collect and analyse information such as that found on websites as well as academic and grey literature, in real-time to create a ‘living review’ of known research activities.