Gravity Current Driven Smoke Dispersion In a Stratified Ambient. Smoke from bushfires transported by gravity currents, and known to occur nationwide, caused the shutdown of businesses, education and events in Canberra in 2019. Recent scientific investigations have shown that the speed of propagation and concentration of smoke in these three-dimensional gravity currents have a long term ‘memory’ of their initial configuration. In this project, high-fidelity computational and experimental techniq ....Gravity Current Driven Smoke Dispersion In a Stratified Ambient. Smoke from bushfires transported by gravity currents, and known to occur nationwide, caused the shutdown of businesses, education and events in Canberra in 2019. Recent scientific investigations have shown that the speed of propagation and concentration of smoke in these three-dimensional gravity currents have a long term ‘memory’ of their initial configuration. In this project, high-fidelity computational and experimental techniques will be used to elucidate the fundamental mechanisms of gravity current entrainment and propagation. This knowledge will set a strong foundation to improve operational forecasts of smoke transport that will allow government agencies to better respond to the negative impact of these complicated flows.Read moreRead less
Approximation theory of structured neural networks . Mathematical theory for deep learning has been desired due to the power applications of deep neural networks to deal with big data in various practical domains. The main difficulty lies in the structures and architectures imposed to networks designed for specific learning tasks. Neither the classical approximation theory nor the recent one for depths of ReLU neural networks can be applied due to the structures imposed for processing large dime ....Approximation theory of structured neural networks . Mathematical theory for deep learning has been desired due to the power applications of deep neural networks to deal with big data in various practical domains. The main difficulty lies in the structures and architectures imposed to networks designed for specific learning tasks. Neither the classical approximation theory nor the recent one for depths of ReLU neural networks can be applied due to the structures imposed for processing large dimensional data such as natural images of tens of thousands of dimensions. This project aims at an approximation theory for structured neural networks. We plan to establish mathematical theories for deconvolution with deep convolutional neural networks, operator learning, and spectral graph networks. Read moreRead less
Human Scheduling of Perceptual Tasks. This project aims to develop a novel approach for synthesising how people prioritise information with theories of attention and decision making. Characterising inefficient scheduling in the tradeoff between the difficulty and the cost/benefit of different subtasks will allow the development of a formal computional model that generalises statistical models of rank order data to a theory of the timing of scheduling decisions and task completions. Outcomes incl ....Human Scheduling of Perceptual Tasks. This project aims to develop a novel approach for synthesising how people prioritise information with theories of attention and decision making. Characterising inefficient scheduling in the tradeoff between the difficulty and the cost/benefit of different subtasks will allow the development of a formal computional model that generalises statistical models of rank order data to a theory of the timing of scheduling decisions and task completions. Outcomes include benchmark data from a novel paradigm for studying perceptual decisions and behavior and a model which can explain and predict human scheduling. This project aims to benefit industry by allowing for the simulation of information prioritisation by human agents in complex environments.Read moreRead less
Large Scale Natural Convection Boundary Layers with Non-Boussinesq Effects. This proposal aims to understand and predict heat transfer by turbulent natural convection in two scenarios, firstly at very large environmental scales, such as occur on melting Antarctic ice sheets, and secondly convection involving very large temperature differences such as occur in solar thermal power plants and industrial processes. These natural convection flow regimes are incredibly difficult to investigate directl ....Large Scale Natural Convection Boundary Layers with Non-Boussinesq Effects. This proposal aims to understand and predict heat transfer by turbulent natural convection in two scenarios, firstly at very large environmental scales, such as occur on melting Antarctic ice sheets, and secondly convection involving very large temperature differences such as occur in solar thermal power plants and industrial processes. These natural convection flow regimes are incredibly difficult to investigate directly but by focusing on the fundamental dynamics of the turbulent flows using large scale numerical simulations and innovative experiments, the project is expected to develop better analytical and computational models which will underpin improvements in
global ocean models and improve energy efficiency.Read moreRead less
Trisections, triangulations and the complexity of manifolds. This project aims at practical representations of 3-dimensional and 4-dimensional spaces as needed in applications. Topology is the mathematical study of the shapes of spaces. Geometry endows spaces with additional structure such as distance, angle and curvature. Special combinatorial structures, such as minimal triangulations, are often closely connected to geometric structures or topological properties. This project aims to construct ....Trisections, triangulations and the complexity of manifolds. This project aims at practical representations of 3-dimensional and 4-dimensional spaces as needed in applications. Topology is the mathematical study of the shapes of spaces. Geometry endows spaces with additional structure such as distance, angle and curvature. Special combinatorial structures, such as minimal triangulations, are often closely connected to geometric structures or topological properties. This project aims to construct computable invariants, connectivity results for triangulations, and algorithms to recognise fundamental topological properties and structures such as trisections and bundles.Read moreRead less
Approximate algorithms and architectures for area efficient system design. This project aims to develop simpler but reliable image recognition systems that can run on low-cost, small-scale platforms, for use in driver monitoring system (DMS) applications. Cheaper reliable DMS will lead to wider availability of this technology to end users and improve safety of motor vehicles. This project will develop approximate algorithmic and circuit techniques, provide training for research students and buil ....Approximate algorithms and architectures for area efficient system design. This project aims to develop simpler but reliable image recognition systems that can run on low-cost, small-scale platforms, for use in driver monitoring system (DMS) applications. Cheaper reliable DMS will lead to wider availability of this technology to end users and improve safety of motor vehicles. This project will develop approximate algorithmic and circuit techniques, provide training for research students and build capability in the area of approximate computing. It is also expected to lead to commercial products, licences and revenue, which will enable new job creation.
Read moreRead less
Scaling laws for aerodynamics of moving wings in the Martian atmosphere. This project aims to increase understanding of the aerodynamics of bio-inspired flight in the low-density atmosphere of Mars. The significance of flight in planetary exploration is shown by the ongoing success of the Ingenuity helicopter on Mars, and the Dragonfly rotorcraft planned for use on Titan. Expected outcomes of this project will be innovative numerical modelling techniques validated using local specially designed ....Scaling laws for aerodynamics of moving wings in the Martian atmosphere. This project aims to increase understanding of the aerodynamics of bio-inspired flight in the low-density atmosphere of Mars. The significance of flight in planetary exploration is shown by the ongoing success of the Ingenuity helicopter on Mars, and the Dragonfly rotorcraft planned for use on Titan. Expected outcomes of this project will be innovative numerical modelling techniques validated using local specially designed low-pressure experimental facilities. Benefits will be more accurate design guidance for efficient and robust flapping and rotary wing robotic vehicles for Mars and other space exploration that take advantage of the unique atmospheric conditions, and in placing Australia at the forefront of such design technology.Read moreRead less
Quantum computation: through the algorithm and complexity theory lens. This project aims to advance our knowledge of quantum computation through the lens of algorithm and complexity theory. Three core areas of the theory will be examined: interactive computing models, query complexity, and circuit lower bounds. The expected outcomes include: revealing the quantum advantages of interactive computing models; techniques for verifying quantum devices in the cloud and quantum cloud computing in gener ....Quantum computation: through the algorithm and complexity theory lens. This project aims to advance our knowledge of quantum computation through the lens of algorithm and complexity theory. Three core areas of the theory will be examined: interactive computing models, query complexity, and circuit lower bounds. The expected outcomes include: revealing the quantum advantages of interactive computing models; techniques for verifying quantum devices in the cloud and quantum cloud computing in general; sharpening the separation between algorithm performance in quantum and classical query models; establishing both unconditional and conditional hardness results for quantum circuits. This comprehensive understanding will enhance Australia's research portfolio in the theory of quantum computing.Read moreRead less
Devising tools for big data sets to support computational movement analysis. This project aims to devise practical fundamental algorithms and multi-purpose data structures with performance guarantees for big spatio-temporal data sets. Systematic analysis of trajectory data has been occurring since the 1950s, but with the recent technological advances the size of the data sets has recently soared. Existing computational tools were developed for small to mid-size data sets. This project aims to d ....Devising tools for big data sets to support computational movement analysis. This project aims to devise practical fundamental algorithms and multi-purpose data structures with performance guarantees for big spatio-temporal data sets. Systematic analysis of trajectory data has been occurring since the 1950s, but with the recent technological advances the size of the data sets has recently soared. Existing computational tools were developed for small to mid-size data sets. This project aims to devise practical fundamental algorithms that will enable the development of domain specific tools for a wide range of applications, including sports, behavioural ecology, transport, and surveillance.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE230100178
Funder
Australian Research Council
Funding Amount
$453,913.00
Summary
Fast, lightweight and live nanopore sequencing analysis. This project aims to address limitations in nanopore sequencing (latest emerging technology in genomics) by applying advanced computational methods. This project expects to create new knowledge in bioinformatics and computer science through innovative approaches that leverage the live data streaming capability of nanopore devices to deliver results rapidly, or in real-time. Expected outcomes include improved, highly efficient analysis meth ....Fast, lightweight and live nanopore sequencing analysis. This project aims to address limitations in nanopore sequencing (latest emerging technology in genomics) by applying advanced computational methods. This project expects to create new knowledge in bioinformatics and computer science through innovative approaches that leverage the live data streaming capability of nanopore devices to deliver results rapidly, or in real-time. Expected outcomes include improved, highly efficient analysis methods and designs for future creation of custom computer hardware for nanopore analysis. This will facilitate widespread adoption of nanopore technology in bioscience research and applied domains (health, agriculture, ecology, biosecurity and forensics), including for portable in-the-field applications. Read moreRead less