Discovery Early Career Researcher Award - Grant ID: DE200101253
Funder
Australian Research Council
Funding Amount
$349,586.00
Summary
Making Machine Learning Fair(er). This project aims to develop and implement statistical methods to fight against algorithm bias. In doing so, this project expects to generate new knowledge in the mathematical sciences by employing innovative and interdisciplinary approaches to the development of fairness constraints on machine learning algorithms. Fairness will be seen through the lens of invariance, allowing the developed conceptual framework to find broad applications. Expected outcomes of t ....Making Machine Learning Fair(er). This project aims to develop and implement statistical methods to fight against algorithm bias. In doing so, this project expects to generate new knowledge in the mathematical sciences by employing innovative and interdisciplinary approaches to the development of fairness constraints on machine learning algorithms. Fairness will be seen through the lens of invariance, allowing the developed conceptual framework to find broad applications. Expected outcomes of this project include improved techniques for imposing invariance on deep learning algorithms. This should provide significant benefits to the general public by contributing to the advancement of socially responsible and conscientious machine learning.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE180100203
Funder
Australian Research Council
Funding Amount
$348,575.00
Summary
Deep space-time models for modelling complex environmental phenomena. This project aims to adapt deep-learning models, used in areas of artificial intelligence such as image tagging and automatic text translation, to improve our understanding of the environment. The project expects to develop new theory for deep-learning models to learn from measurement data and numerical-model output about environmental phenomena that evolve in space and time, such as ice sheets and the atmosphere. Expected out ....Deep space-time models for modelling complex environmental phenomena. This project aims to adapt deep-learning models, used in areas of artificial intelligence such as image tagging and automatic text translation, to improve our understanding of the environment. The project expects to develop new theory for deep-learning models to learn from measurement data and numerical-model output about environmental phenomena that evolve in space and time, such as ice sheets and the atmosphere. Expected outcomes include the ability to provide reliable predictions and quantification of uncertainty on environmental concerns of national importance, such as sea-level rise. Key benefits include improved risk management and mitigation, for example through financial incentives or infrastructure planning.Read moreRead less
Industrial Transformation Training Centres - Grant ID: IC190100031
Funder
Australian Research Council
Funding Amount
$3,973,202.00
Summary
ARC Training Centre in Data Analytics for Resources and Environments (DARE). Understanding the cumulative impact of actions regarding the use of our resources has important long-term consequences for Australia’s economic, societal and environmental health. Yet despite the importance of these cumulative impacts, and the availability of data, many decisions and policies are based on limited amounts of data and rudimentary data analysis, with little appreciation of the critical role that understand ....ARC Training Centre in Data Analytics for Resources and Environments (DARE). Understanding the cumulative impact of actions regarding the use of our resources has important long-term consequences for Australia’s economic, societal and environmental health. Yet despite the importance of these cumulative impacts, and the availability of data, many decisions and policies are based on limited amounts of data and rudimentary data analysis, with little appreciation of the critical role that understanding and quantifying uncertainty plays in the process. The aim of Data Analytics in Resources and Environment (DARE) is to develop and deliver the data science skills and tools for Australia’s resource industries to make the best possible evidence-based decisions in exploiting and stewarding the nation’s natural resources.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE200100200
Funder
Australian Research Council
Funding Amount
$418,398.00
Summary
Next generation causal inference methods for biological data. This project aims to develop next generation causal inference methods for analysing biological data especially the single cell sequencing data and their applications in cell biology. Although Artificial Intelligence and Statistical Machine Learning have been applied successfully in many fields, including biological research, there is still a serious lack of methods for interpreting and reasoning about the mechanism of biological syste ....Next generation causal inference methods for biological data. This project aims to develop next generation causal inference methods for analysing biological data especially the single cell sequencing data and their applications in cell biology. Although Artificial Intelligence and Statistical Machine Learning have been applied successfully in many fields, including biological research, there is still a serious lack of methods for interpreting and reasoning about the mechanism of biological systems, the ultimate goal of research in many areas. Efficient data-driven causality discovery approaches developed by the project will be a timely and significant contribution to the knowledge of biology and statistics as well as the battle against health threats.
Read moreRead less
Generalised Degrees of Freedom and Probabilistic Regularisation. This project intends to develop novel statistical tools for more accurate prediction by taking account of model complexity and uncertainties associated with the fitting procedure. The project also plans to develop a novel shrinkage approach via new penalty functions to avoid over-fitting and asymptotic properties. The key applications may include genetic studies where the number of predictors is large and biological experiments whe ....Generalised Degrees of Freedom and Probabilistic Regularisation. This project intends to develop novel statistical tools for more accurate prediction by taking account of model complexity and uncertainties associated with the fitting procedure. The project also plans to develop a novel shrinkage approach via new penalty functions to avoid over-fitting and asymptotic properties. The key applications may include genetic studies where the number of predictors is large and biological experiments where multivariate and temporal data are often collected – for example economical breeding in animal and fish farming and more effectively detecting the genes of interest in genetic studies on human, animals and plants.Read moreRead less
Novel statistical methods for data with non-Euclidean geometric structure. This project aims to develop new flexible regression models and classification algorithms, along with robust and efficient inference methods, applicable to a wide range of non-Euclidean data types which arise in many fields of science, business and technology. There are serious flaws with currently available methods of analysis for non-Euclidean data. This project expects to transform such analyses by providing new quanti ....Novel statistical methods for data with non-Euclidean geometric structure. This project aims to develop new flexible regression models and classification algorithms, along with robust and efficient inference methods, applicable to a wide range of non-Euclidean data types which arise in many fields of science, business and technology. There are serious flaws with currently available methods of analysis for non-Euclidean data. This project expects to transform such analyses by providing new quantitative tools within a unifying framework. The anticipated project outcomes will be of mathematical interest and valuable in applications such as finance (predicting Australian stock returns); modelling electroencephalography data; Australian geochemical data, relating to sediments; and Australian X-ray tumour image data. Read moreRead less
High Predictive Performance Models via Semi-Parametric Survival Regression. This project will develop novel statistical models for high prediction performance. When applied to help doctor to treat patients, these models allow the users to include gene or other biomarkers for predicting effectiveness of a treatment. When applied to risk management in finance, these models are capable to include an organization's or individual's ongoing finance status to predict, for example, the probability of or ....High Predictive Performance Models via Semi-Parametric Survival Regression. This project will develop novel statistical models for high prediction performance. When applied to help doctor to treat patients, these models allow the users to include gene or other biomarkers for predicting effectiveness of a treatment. When applied to risk management in finance, these models are capable to include an organization's or individual's ongoing finance status to predict, for example, the probability of or time to loan default. Innovative computational methods will be developed for fitting these models. Compared to traditional prediction method, this approach allows greater flexibility while being superior in terms of statistical accuracy and bias. Extensive analyses of healthcare data from diverse fields will be undertaken.Read moreRead less
Precision ecology: the modern era of designed experiments in plant ecology. This project aims to develop the field of precision ecology, forging a new era of designed experiments where sampling is informed by research questions and what is known about the ecological process being studied. Through the development of novel statistical methods, new experiments globally will be designed to answer important ecological questions including what influence abiotic and biotic factors have on plant commun ....Precision ecology: the modern era of designed experiments in plant ecology. This project aims to develop the field of precision ecology, forging a new era of designed experiments where sampling is informed by research questions and what is known about the ecological process being studied. Through the development of novel statistical methods, new experiments globally will be designed to answer important ecological questions including what influence abiotic and biotic factors have on plant communities over time and different spatial scales. Expected outcomes include new methods and tools that will modernise how future experiments will be conducted in plant ecology. This will provide significant transdisciplinary benefits including new statistical methods that target scientific discovery in ecological studies.Read moreRead less
Principled statistical methods for high-dimensional correlation networks. This project aims to develop a novel and principled approach for building correlation networks. Correlation networks aim to identify the most significant associations present in modern massive datasets, and have numerous applications, ranging from the biomedical and environmental sciences to the social sciences. Nodes of such networks represent features, and edges represent associations, or the lack thereof. Current method ....Principled statistical methods for high-dimensional correlation networks. This project aims to develop a novel and principled approach for building correlation networks. Correlation networks aim to identify the most significant associations present in modern massive datasets, and have numerous applications, ranging from the biomedical and environmental sciences to the social sciences. Nodes of such networks represent features, and edges represent associations, or the lack thereof. Current methods are not readily scalable to modern ultra-high dimensional settings, and do not account for uncertainty in the estimated associations. This project will develop a principled, highly scalable methodology for building such networks, which incorporates uncertainty quantification. Emphasis is placed on modern ultra-high dimensional settings in which differentiating a true correlation from a spurious one is a notoriously difficult task.Read moreRead less
Bayesian inversion and computation applied to atmospheric flux fields. This project aims to make use of unprecedented sources of measurements, from remote sensing and in situ data, to estimate the sources and sinks of greenhouse gases. An overabundance of greenhouse gases in Earth's atmosphere is arguably the most serious long-term threat to the planet's ecosystems. This project will combine measurement uncertainties, process uncertainties in the physical transport models, and any parameter unce ....Bayesian inversion and computation applied to atmospheric flux fields. This project aims to make use of unprecedented sources of measurements, from remote sensing and in situ data, to estimate the sources and sinks of greenhouse gases. An overabundance of greenhouse gases in Earth's atmosphere is arguably the most serious long-term threat to the planet's ecosystems. This project will combine measurement uncertainties, process uncertainties in the physical transport models, and any parameter uncertainties, to provide reliable uncertainty quantification for the estimates. This will be achieved with new Bayesian spatio-temporal inversions and big-data computational strategies. The resulting statistical inferences on greenhouse-gas flux fields will enable the development of critical mitigation strategies. These new statistical inferences will be a valuable resource to policy-makers worldwide, who are assessing progress towards global commitments. Further, the final product may assist in developing cost-effective mitigation strategies in the presence of uncertainty.Read moreRead less