Alexander G. de G. Matthews
I am a PhD student in the Machine Learning Group supervised by Zoubin Ghahramani. As an undergraduate I studied Natural Sciences at the University of Cambridge, specialising in theoretical physics. My fourth year project, which was later published, studied scattering in the fractional quantum Hall effect with Nigel Cooper. After that I worked in industry for Navetas Energy Management, a University of Oxford spin-out company which applies machine learning to the problem of home energy disaggregation.
I am interested in Bayesian nonparametric machine learning, particularly spatial point processes and Gaussian processes. I am also interested in approximate Bayesian inference, particularly Markov chain Monte Carlo methods, belief propagation and variational methods.
My AISTATS 2015 paper with James Hensman and Zoubin Ghahramani entitled “Scalable Variational Gaussian Process Classification” can be found here:
I presented a workshop paper written with James Hensman and Zoubin Ghahramani on variational mixture distributions at the NIPS variational inference workshop 2014. The paper and slides for my talk are available.
An arXiv pre-print of my work with James Hensman, Maurizio Filippone and Zoubin Ghahramani combining Markov chain Monte Carlo with sparse variational methods for Gaussian processes can be found at:
An arXiv pre-print of my work with with James Hensman, Richard Turner and Zoubin Ghahramani interpreting Titsias’ sparse variational framework as a Kullback-Leibler divergence between approximating and posterior processes can be found at:
An arXiv pre-print of my work on spatial point point processes with Zoubin Ghahramani can be found at: