Discovery Early Career Researcher Award - Grant ID: DE220100265
Funder
Australian Research Council
Funding Amount
$417,000.00
Summary
A closed-loop human–agent learning framework to enhance decision making. This project aims to design a foundational human–agent learning framework to augment the decision making process, using reinforcement and closed-loop mechanisms to enable symbiosis between a human and an artificial-intelligence agent. It envisages significant new technologies to promote controllability and efficient and safe exploration of an environment for decision actions – drastically boosting learning effectiveness and ....A closed-loop human–agent learning framework to enhance decision making. This project aims to design a foundational human–agent learning framework to augment the decision making process, using reinforcement and closed-loop mechanisms to enable symbiosis between a human and an artificial-intelligence agent. It envisages significant new technologies to promote controllability and efficient and safe exploration of an environment for decision actions – drastically boosting learning effectiveness and interpretability in decision making. Expected outcomes will benefit national cybersecurity by improving our understanding of vulnerabilities and threats involving decision actions, and by ensuring that human feedback and evaluations can help prevent catastrophic events in explorations of dynamic and complex environments.Read moreRead less
Dissecting The Pseudoexfoliation Syndrome With Complementary Genetic, Proteomic And Biophysical Strategies
Funder
National Health and Medical Research Council
Funding Amount
$490,352.00
Summary
Pseudoexfoliation syndrome (PEX) is an eye condition in which flaky material deposits in the eye, greatly increasing the risk of cataract and glaucoma which can lead to blindness. PEX is also associated with heart disease, strokes and aneurysms. Cataract surgery in PEX patients has a higher rate of complications. In this project we will determine the nature of PEX material and why it forms. This knowlege will facilitate better diagnosis and treatment of PEX preventing associated blindness.
Industrial Transformation Research Hubs - Grant ID: IH180100002
Funder
Australian Research Council
Funding Amount
$5,000,000.00
Summary
ARC Research Hub for Driving Farming Productivity and Disease Prevention. The ARC Research Hub for Driving Farming Productivity and Disease Prevention aims to increase farm production and disease prevention through advancing and transferring new artificial intelligence technologies into industrial deployment. The Hub will combine machine vision, machine learning, software quality control, engineering, biology, and farming industries to develop technologies to build more intelligent systems. Thes ....ARC Research Hub for Driving Farming Productivity and Disease Prevention. The ARC Research Hub for Driving Farming Productivity and Disease Prevention aims to increase farm production and disease prevention through advancing and transferring new artificial intelligence technologies into industrial deployment. The Hub will combine machine vision, machine learning, software quality control, engineering, biology, and farming industries to develop technologies to build more intelligent systems. These dynamic systems will help determine what goal to achieve and the most efficient plan to achieve it. This Hub is expected to contribute to higher farming efficiency, lower production costs and fewer disease risks, giving the Australian industry new business opportunities and an international competitive advantage.Read moreRead less
Visual analytics for massive multivariate networks. Visual analytics for massive multivariate networks. This project aims to create methods to visually analyse massive multivariate networks. The amount of network data available has exploded in recent years: software systems, social networks and biological systems have millions of nodes and billions of edges with multivariate attributes. Their size and complexity makes these data sets hard to exploit. More efficient ways to understand the data ar ....Visual analytics for massive multivariate networks. Visual analytics for massive multivariate networks. This project aims to create methods to visually analyse massive multivariate networks. The amount of network data available has exploded in recent years: software systems, social networks and biological systems have millions of nodes and billions of edges with multivariate attributes. Their size and complexity makes these data sets hard to exploit. More efficient ways to understand the data are needed. This project will design, implement and evaluate visualisation methods for massive multivariate network data sets. This research is expected to be used by Australian software development, biotechnology and security companies to exploit their data.Read moreRead less
A Genome-wide Association Study In 2000 Glaucoma Cases With Matched Controls Using Equimoloar DNA Pools
Funder
National Health and Medical Research Council
Funding Amount
$610,267.00
Summary
Glaucoma is a common cause of loss of vision worldwide but we are unable to predict which people are at high risk of blindness. We aim to discover the genetic risk factors for glaucoma. We will use cutting edge genetic technology to assess the whole genome in thousands of patients with glaucoma. We hope to identify important new glaucoma genes, which could lead to the development of diagnostic tests and treatments which will provide the most cost-efficient ways to prevent glaucoma blindness.
A Genome-wide Association Scan To Identify Genetic Risk Factors For Sight Threatening Diabetic Retinopathy
Funder
National Health and Medical Research Council
Funding Amount
$982,203.00
Summary
Diabetic eye disease is an important complication of diabetes that can lead to blindness. Very little is known about how diabetes causes eye disease, but genetics is known to play a role. We aim to identify genes that contribute to eye disease in diabetes patients. We will compare genes between patients with diabetes with and without severe diabetic eye disease using cutting edge genomic technology. We hope to be able to better predict risk of blindness and to move towards novel treatments.
Building Australia's next-generation ocean-sea ice model. Ocean and sea ice models are used for predicting future ocean and climate states, and for climate process research. This project aims to bring the next generation of ocean-sea ice models to Australia and configure the models for our local priorities. The ultimate goal is to create a new coupled ocean-sea ice model for Australia that includes surface waves and biogeochemistry. The model will be optimised and evaluated on Australian facilit ....Building Australia's next-generation ocean-sea ice model. Ocean and sea ice models are used for predicting future ocean and climate states, and for climate process research. This project aims to bring the next generation of ocean-sea ice models to Australia and configure the models for our local priorities. The ultimate goal is to create a new coupled ocean-sea ice model for Australia that includes surface waves and biogeochemistry. The model will be optimised and evaluated on Australian facilities, and released for community use. These developments underpin future ocean state forecasts, sea ice forecasts, wave forecasts, decadal climate prediction and climate process studies. The project will benefit search and rescue, Defence and shipping operations, and will enhance future climate projections.Read moreRead less
Developing a Unified Theory of Episodic Memory. This project aims to develop a model of episodic memory and to apply the model to both adult and child development data. Unlike current approaches, the model is expected to address multiple memory tasks including item recognition, associative recognition, source recognition and cued recall, and also aims to address reaction time data, allowing different sources of interference causing forgetting in adults to be identified. By addressing both encodi ....Developing a Unified Theory of Episodic Memory. This project aims to develop a model of episodic memory and to apply the model to both adult and child development data. Unlike current approaches, the model is expected to address multiple memory tasks including item recognition, associative recognition, source recognition and cued recall, and also aims to address reaction time data, allowing different sources of interference causing forgetting in adults to be identified. By addressing both encoding and retrieval processes, the model can assess how changes in different sources of interference modulate performance through the trajectory of early development. Hierarchical Bayesian estimation aims to enable a simultaneous account of multiple tasks and support future deployment in applied contexts.Read moreRead less
Automated internet warnings to prevent viewing of minor-adult sex images. Since the advent of the internet and digital cameras, the market for child exploitation material (CEM) has boomed. This project aims to explore how the visual appearance of warning messages influences internet users. It plans to conduct a randomised controlled experiment with naïve participants on a real-life website to test the effectiveness of messages designed to discourage viewers of legal ‘barely legal’ pornography. I ....Automated internet warnings to prevent viewing of minor-adult sex images. Since the advent of the internet and digital cameras, the market for child exploitation material (CEM) has boomed. This project aims to explore how the visual appearance of warning messages influences internet users. It plans to conduct a randomised controlled experiment with naïve participants on a real-life website to test the effectiveness of messages designed to discourage viewers of legal ‘barely legal’ pornography. It is anticipated that results will assist policing efforts by indicating whether warnings can be used to dissuade first-time CEM viewers and whether differences exist between harm or deterrent-focused messages.Read moreRead less
Choice models for learning and memory. Life is filled with familiar choices that often require quick decisions about objects in the environment and the contents of memory. This project examines how we learn to make rapid and accurate choices and how we quickly asses the level of confidence we have in recognition decisions based on our memories.