Deep Learning that Scales. Deep learning has dramatically improved the accuracy of a breathtaking variety of tasks in AI such as image understanding and natural language processing. This project addresses fundamental bottlenecks when attempting to develop deep learning applications at scale. First, this project proposes efficient neural architecture search that is orders of magnitude faster than previously reported, abstracting away the most complex part of deep learning. Second, we will desig ....Deep Learning that Scales. Deep learning has dramatically improved the accuracy of a breathtaking variety of tasks in AI such as image understanding and natural language processing. This project addresses fundamental bottlenecks when attempting to develop deep learning applications at scale. First, this project proposes efficient neural architecture search that is orders of magnitude faster than previously reported, abstracting away the most complex part of deep learning. Second, we will design very efficient binary networks, enabling large-scale deployment of deep learning to mobile devices. Thus this project will overcome two primary limitations of deep learning generally, however, and will greatly increase its already impressive domain of practical application.Read moreRead less
Adapting Deep Learning for Real-world Medical Image Datasets. The project aims to investigate new deep learning modelling approaches to leverage real-world large-scale image data sets that contain noisy and incomplete labels and imbalanced class prevalence – to enable the use of these data sets for modelling deep learning classifiers. Expected outcomes include an innovative method for modelling deep learning classifiers. The research will involve new inter-disciplinary and international collabor ....Adapting Deep Learning for Real-world Medical Image Datasets. The project aims to investigate new deep learning modelling approaches to leverage real-world large-scale image data sets that contain noisy and incomplete labels and imbalanced class prevalence – to enable the use of these data sets for modelling deep learning classifiers. Expected outcomes include an innovative method for modelling deep learning classifiers. The research will involve new inter-disciplinary and international collaborations with machine learning and medical image analysis research institutions. This should provide significant benefits, such as better understanding of deep learning theory, new deep learning applications that can use previously unexplored data sets, and training for the future Australian workforce.Read moreRead less
Automated analysis of multi-modal medical data using deep belief networks. This project will develop an improved breast cancer computer-aided diagnosis (CAD) system that incorporates mammography, ultrasound and magnetic resonance imaging. This system will be based on recently developed deep learning techniques, which have the capacity to process multi-modal data in a unified and optimal manner. The advantage of this technique is that it is able to automatically learn both the relevant features t ....Automated analysis of multi-modal medical data using deep belief networks. This project will develop an improved breast cancer computer-aided diagnosis (CAD) system that incorporates mammography, ultrasound and magnetic resonance imaging. This system will be based on recently developed deep learning techniques, which have the capacity to process multi-modal data in a unified and optimal manner. The advantage of this technique is that it is able to automatically learn both the relevant features to analyse in each modality and the hidden relationships between them. The use of deep belief networks has produced promising results in several fields, such as speech recognition, and so this project believes that our approach has the potential to improve both the sensitivity and specificity of breast cancer detection.Read moreRead less
From insects to robots: how brains make predictions and ignore distractions. This project aims to address fundamental questions in neuroscience and to integrate this biological understanding with the development of leading-edge robotics. Whether a human catching a ball or a dragonfly feeding in a swarm, brains have the remarkable ability to predict the future location of moving targets. The brain predicts in the presence of distractions and even if the target disappears, for example, when hidden ....From insects to robots: how brains make predictions and ignore distractions. This project aims to address fundamental questions in neuroscience and to integrate this biological understanding with the development of leading-edge robotics. Whether a human catching a ball or a dragonfly feeding in a swarm, brains have the remarkable ability to predict the future location of moving targets. The brain predicts in the presence of distractions and even if the target disappears, for example, when hidden behind another object. This project will investigate how brains use both environmental and internal information to select a target and predict its future location. By implementing bio-inspired computations in hardware, this project aims to provide significant benefits such as improving autonomous systems for defence, health and transportation.Read moreRead less
Deep reinforcement learning for discovering and visualising biomarkers. This project aims to develop novel methods for discovering and visualising optimal bio-markers from chest computed tomography images based on extensions of recently developed deep reinforcement learning techniques. The extensions proposed in this project will advance medical image analysis by allowing an efficient analysis of large dimensionality inputs in their original high resolution. In addition, this project will be the ....Deep reinforcement learning for discovering and visualising biomarkers. This project aims to develop novel methods for discovering and visualising optimal bio-markers from chest computed tomography images based on extensions of recently developed deep reinforcement learning techniques. The extensions proposed in this project will advance medical image analysis by allowing an efficient analysis of large dimensionality inputs in their original high resolution. In addition, this project will be the first approach capable of discovering previously unknown biomarkers associated with important clinical outcomes. The project will validate the approach on a real-world case study data set concerning the prediction of five-year survival of chronic disease.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE150100548
Funder
Australian Research Council
Funding Amount
$359,000.00
Summary
Neural and robotic correlates of predictive coding and selective attention. Whether a human catching a ball, a dog leaping at a frisbee or a dragonfly hunting prey amidst a swarm, brains both large and small have evolved the ability to focus attention on one moving target, even in the presence of distracters. This project aims to investigate how brains solve this challenging problem by recording the activity of dragonfly neurons that selectively attend to one target whilst ignoring others. The p ....Neural and robotic correlates of predictive coding and selective attention. Whether a human catching a ball, a dog leaping at a frisbee or a dragonfly hunting prey amidst a swarm, brains both large and small have evolved the ability to focus attention on one moving target, even in the presence of distracters. This project aims to investigate how brains solve this challenging problem by recording the activity of dragonfly neurons that selectively attend to one target whilst ignoring others. The project aims to examine how expectation and attention are encoded in the brain and will build an autonomous robot using computational models bio-inspired from this neuronal processing. Robots capable of visually perceiving and interacting with targets in natural environments have applications in health, surveillance and defence.Read moreRead less
Multi-modal virtual microscopy for quantitative diagnostic pathology. This project will contribute to the next generation of virtual microscopy systems that provide innovative features capable of significantly increasing the adoption of digital imaging technology throughout the field of diagnostic pathology. These tools will especially contribute to the screening and diagnosis of cervical, lung and bladder cancer.