I've also worked on two Kaggle competitions in my spare time: the AES seizure prediction challenge and the first National Data Science Bowl. In both, our team name was Neuroglycerin, and in both we achieved scores in the top 10%.
In addition, I've worked on setting up an early-stage Jupyter platform for students at the University of Edinburgh to use in the course Data, Design and Society. The server integrates JupyterHub with the EASE authentication system and runs in docker.
PhD Neuroinformatics, Edinburgh University - 2014 to present
Supervised by Amos Storkey and D K Arvind, as part of the Bayeswatch group, my project is focused on resource efficiency and architecture search in machine learning. My research interests are stochastic variational methods, Bayesian deep learning, network compression, probabilistic programming, Bayesian optimisation and open science. Recent publications include:
- "Moonshine: Distilling with Cheap Convolutions" at NIPS 2018
- "Resource-Efficient Feature Gathering at Test Time", accepted at Reliable Machine Learning in the Wild, NIPS 2016.
In the process of my PhD I have replicated the work of other papers in the field, such as "Variational Dropout and the Local Reparameterization Trick", "Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization" and "The Shattered Gradients Problem". A more complete list can be found here and my GitHub profile is here. I've also posted reviews of papers on Short Science.
I'm proficient in: Python, C, Verilog, Matlab, Tensorflow, Theano, PyTorch, Lasagne, Bash scripting and Linux.
I developed a dockerised JupyterHub deployment for the Data, Design and Society course to provide students with a fully configured web coding environment.
I competed in the AES Seizure Prediction Challenge (16th/504) and the National Data Science Bowl (57th/1049) in the team Neuroglycerin. The code written for both competitions is available: hail-seizure and neukrill-net.
MRes Neuroinformatics, Edinburgh University - 2013 to 2014
This programme was designed as a precursor to a three year PhD project at the Doctoral Training Centre in Neuroinformatics and Computational Neuroscience, providing background in neuroscience and machine learning.
Master's Project - "Weighting Protein-Protein Interaction Networks"
MEng Electrical Engineering and Electronics, Edinburgh University - 2008 to 2013
Graduated in 2013 - First class with honours.
This CV is also available as a pdf.