6 papers
Not Half Bad: Exploring Half-Precision in Graph Convolutional Neural Networks
John Brennan, Stephen Bonner, Amir Atapour-Abarghouei +3
With the growing significance of graphs as an effective representation of data in numerous applications, efficient graph analysis using modern machine learning is receiving a growi…
Camera Bias in a Fine Grained Classification Task
Philip T. Jackson, Stephen Bonner, Ning Jia +3
We show that correlations between the camera used to acquire an image and the class label of that image can be exploited by convolutional neural networks (CNN), resulting in a mode…
Temporal Neighbourhood Aggregation: Predicting Future Links in Temporal Graphs via Recurrent Variational Graph Convolutions
Stephen Bonner, Amir Atapour-Abarghouei, Philip T Jackson +5
Graphs have become a crucial way to represent large, complex and often temporal datasets across a wide range of scientific disciplines. However, when graphs are used as input to ma…
Combining Mathematical Morphology and the Hilbert Transform for Fully Automatic Nuclei Detection in Fluorescence Microscopy
Carl J. Nelson, Philip T. G. Jackson, Boguslaw Obara
Accurate and reliable nuclei identification is an essential part of quantification in microscopy. A range of mathematical and machine learning approaches are used but all methods h…
Phenotypic Profiling of High Throughput Imaging Screens with Generic Deep Convolutional Features
Philip T. Jackson, Yinhai Wang, Sinead Knight +5
While deep learning has seen many recent applications to drug discovery, most have focused on predicting activity or toxicity directly from chemical structure. Phenotypic changes e…
Style Augmentation: Data Augmentation via Style Randomization
Philip T. Jackson, Amir Atapour-Abarghouei, Stephen Bonner +2
We introduce style augmentation, a new form of data augmentation based on random style transfer, for improving the robustness of convolutional neural networks (CNN) over both class…