The early applications of my work are in the resource constrained inference for embedded and wearable devices. My long-term aim is to develop and popularise my concept of Rational Automated Machine Learner – a learning paradigm whereby a wider set of resource costs and preference considerations can be consulted along with the model accuracy in the AutoML process.
I am, also, a co-founder and the director of the Cambridge Neural Group. CNG works towards making deep learning widely accessible to the progessional public in the UK and abroad.
For more information about my research and teaching activities see the appropriate tabs above.