Partially Observable MDPs, Monte Carlo Methods, and Sustainable Fisheries. Partially Observable Markov Decision Processes (POMDPs) provide a general mathematical framework for sequential decision making under uncertainty. However, solving POMDPs effectively under realistic assumptions remains a challenging problem. This project aims to develop new efficient Monte Carlo algorithms to significantly advance the application of POMDPs to real-world decision problems involving complex action spaces an ....Partially Observable MDPs, Monte Carlo Methods, and Sustainable Fisheries. Partially Observable Markov Decision Processes (POMDPs) provide a general mathematical framework for sequential decision making under uncertainty. However, solving POMDPs effectively under realistic assumptions remains a challenging problem. This project aims to develop new efficient Monte Carlo algorithms to significantly advance the application of POMDPs to real-world decision problems involving complex action spaces and system dynamics. Both theoretical and algorithmic approaches will be applied to sustainable fishery management --- an important problem for Australia and an ideal context for POMDPs. The project will advance research in artificial intelligence, dynamical systems, and fishery operations, and benefit the national economy.Read moreRead less
Dynamic Deep Learning for Electricity Demand Forecasting. This project aims at developing a deep learning technology for high resolution electricity demand forecasting and residential demand response modelling. Electricity consumption data are dynamic and highly uncertain. The deep learning technology expects to provide accurate demand forecasting, and thus enabling optimal use of existing
grid assets and guiding future investments. The expected outcome can support data-driven decision-making in ....Dynamic Deep Learning for Electricity Demand Forecasting. This project aims at developing a deep learning technology for high resolution electricity demand forecasting and residential demand response modelling. Electricity consumption data are dynamic and highly uncertain. The deep learning technology expects to provide accurate demand forecasting, and thus enabling optimal use of existing
grid assets and guiding future investments. The expected outcome can support data-driven decision-making in Australia's electricity distribution network planning and operation by considering future challenges such as integrating battery storage and electric vehicles into the grid, and thus providing reliable energy. The project expects to train next generation expert workforce for Australia's future power grid.Read moreRead less
Interpretable Behaviour Analysis with External Structured Knowledge. This project aims to develop novel interpretable neural models for predictive analytics tasks on human behaviour, operating on sequence behaviour data associated with external supportive structured knowledge. It is expected to present theoretical foundations for robust representation learning on heterogeneous behaviour data and interpretable machine reasoning models, which can support a broad scope of intelligent systems. Expec ....Interpretable Behaviour Analysis with External Structured Knowledge. This project aims to develop novel interpretable neural models for predictive analytics tasks on human behaviour, operating on sequence behaviour data associated with external supportive structured knowledge. It is expected to present theoretical foundations for robust representation learning on heterogeneous behaviour data and interpretable machine reasoning models, which can support a broad scope of intelligent systems. Expected outcomes will be a next-generation interpretable behaviour analysis system with versatile abilities to reason over various data structures and provide a high-level interpretability about its reasoning procedure. The benefits will span the research and industry sectors, e.g., retail, healthcare, service provider.Read moreRead less
Developing key vision technology for automation of aquaculture factory. This project aims to investigate structural, coloured textural, and hyperspectral analysis approaches to achieve automated lobster molt-cycle staging and classification to the level required for commercial production. High labour cost, water contamination, and disease transmission are major barriers in Australian bay lobster aquaculture inhibiting its large scale production. Automation of the production process and reducing ....Developing key vision technology for automation of aquaculture factory. This project aims to investigate structural, coloured textural, and hyperspectral analysis approaches to achieve automated lobster molt-cycle staging and classification to the level required for commercial production. High labour cost, water contamination, and disease transmission are major barriers in Australian bay lobster aquaculture inhibiting its large scale production. Automation of the production process and reducing the human contact with animals are of high priority in the development of this Australian-led emerging industry. The project aims to develop technology to bring this world- first aquaculture factory to large scale production, and create new export opportunities for lobsters and production systems.Read moreRead less
Constraint-based Reasoning for Multi-agent Pathfinding. Automation is a transformative technology for logistics -- using robots to manipulate inventory allows warehouses to be more efficient, and larger-scale, than ever before. But doing this in practice requires efficient, reliable methods for coordinating ever-larger fleets of robots. These problems are extremely difficult, and current approaches either scale poorly or give weak or no guarantees on solution quality. The project will develop t ....Constraint-based Reasoning for Multi-agent Pathfinding. Automation is a transformative technology for logistics -- using robots to manipulate inventory allows warehouses to be more efficient, and larger-scale, than ever before. But doing this in practice requires efficient, reliable methods for coordinating ever-larger fleets of robots. These problems are extremely difficult, and current approaches either scale poorly or give weak or no guarantees on solution quality. The project will develop transformative approaches to multi-agent pathfinding which can handle industrial size problems, and handle all of the complications that arise in practical applications. This will deliver improved cost-effectiveness and productivity to automated warehouse logistics and other agent coordination problems.Read moreRead less
New Insights on Modelling Time Trends with Panel Data: Theory and Practice. This project aims to tackle important challenges in time trend modelling by taking advantage of panel data structures. This project expects to propose flexible models in time trend modelling to retrieve reliable inference. The expected outcomes include innovative econometric models and methods that have a wide range of applications, and are particularly suited for empirical problems within large and complex systems. This ....New Insights on Modelling Time Trends with Panel Data: Theory and Practice. This project aims to tackle important challenges in time trend modelling by taking advantage of panel data structures. This project expects to propose flexible models in time trend modelling to retrieve reliable inference. The expected outcomes include innovative econometric models and methods that have a wide range of applications, and are particularly suited for empirical problems within large and complex systems. This will provide significant benefits to all fields in which data displays any form of trending behaviour. The proposed model is used to evaluate the economic consequences of climate change and global housing market contagion, which provide strong evidence-based insights to the environmental and economic policies in Australia.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE160100568
Funder
Australian Research Council
Funding Amount
$360,000.00
Summary
Towards reliability in combinatorial optimisation. This project intends to develop techniques to ensure that the solutions reported by optimisation tools are correct and verifiable. Combinatorial optimisation problems, where the best solution must be found from a vast set of possibilities, are central to critical sectors of the economy, including shipping, transit, mining and emergency response. Automated tools for these problems can now solve large industrial examples, however, they are incredi ....Towards reliability in combinatorial optimisation. This project intends to develop techniques to ensure that the solutions reported by optimisation tools are correct and verifiable. Combinatorial optimisation problems, where the best solution must be found from a vast set of possibilities, are central to critical sectors of the economy, including shipping, transit, mining and emergency response. Automated tools for these problems can now solve large industrial examples, however, they are incredibly complex artefacts which are prone to error and difficult to test. New methods for ensuring the correctness of automated tools would allow users to trust that the results returned by these tools are correct when making critical decisions.Read moreRead less
Harnessing Imaging And IT Strategies To Expedite Targeted Treatment And Improve Outcomes In Cerebrovascular Diseases
Funder
National Health and Medical Research Council
Funding Amount
$2,914,215.00
Summary
This project will expand on my 25+ years of research in combining neuroimaging methods such as CT and MRI with nascent software tools to better target and coordinate treatment and achieve improved outcomes in cerebrovascular diseases such as stroke. We will develop & improve new CT and MRI methods and leverage latest advances in computer science, such as deep learning and mobile phone app technology, to achieve faster and more accurate identification of patients who can benefit from treatment.
Discovery Early Career Researcher Award - Grant ID: DE210101297
Funder
Australian Research Council
Funding Amount
$429,000.00
Summary
A novel, dictionary-free, multi-contrast MRI method for microscopic imaging. This project aims to develop a novel quantitative imaging technique for comprehensive in vitro and in vivo tissue characterisation on the microscopic scale. The technology innovated in the project could revolutionise microscopic imaging techniques by breaking through the sub-millimetre image resolution bottleneck of current magnetic resonance imaging (MRI) methods. This project expects to generate new knowledge in the e ....A novel, dictionary-free, multi-contrast MRI method for microscopic imaging. This project aims to develop a novel quantitative imaging technique for comprehensive in vitro and in vivo tissue characterisation on the microscopic scale. The technology innovated in the project could revolutionise microscopic imaging techniques by breaking through the sub-millimetre image resolution bottleneck of current magnetic resonance imaging (MRI) methods. This project expects to generate new knowledge in the emerging field of biological imaging and to deliver an integrated imaging platform for mapping various tissue microscopic components at the cellular level. Successful outcomes have the potential for commercialisation and will accelerate a range of fundamental science and engineering studies requiring imaging techniques.Read moreRead less
Searching for near-exact protein models. This project aims to develop novel and efficient heuristic-based algorithms leading to near accurate protein tertiary structure models. Knowledge about protein structures is fundamental to our understanding of living systems. The progress on experimental determination of these structures has been extremely limited and remains an open challenge in molecular biology. Computational prediction of protein structures from sequences is emerging as a promising ap ....Searching for near-exact protein models. This project aims to develop novel and efficient heuristic-based algorithms leading to near accurate protein tertiary structure models. Knowledge about protein structures is fundamental to our understanding of living systems. The progress on experimental determination of these structures has been extremely limited and remains an open challenge in molecular biology. Computational prediction of protein structures from sequences is emerging as a promising approach, but its accuracy is far from satisfactory. The software systems developed in this project will be used in structural identification of target proteins in drug design. This will make drug design process more efficient, saving time and cost, potentially saving lives.Read moreRead less