Corpus Christi College Oxford

Follow Corpus Christi College Oxford on Facebook

Dr Louis Aslett


louis.aslett@stats.ox.ac.uk     


Biography  


I graduated with a BA in Mathematics from Trinity College Dublin and continued to complete a PhD in Statistics at the end of 2012 under the supervision of Simon Wilson, also at Trinity College Dublin.  Before entering research I was Founder and Technical Director of 6 Internet Limited, a server hosting and application development specialist.


Research Interests


I am a postdoctoral research associate on an EPSRC grant in the Department of Statistics at the University of Oxford. This multi-institution research programme spans the universities of Oxford, Warwick, Bristol and Lancaster, with the objective of developing computational and methodological approaches for tackling complex likelihood based inference problems. Specifically, these are problems where calculation of a probability model may be prohibitively slow, or even impossible. Such models arise throughout the physical, life and social sciences.


At present I have three primary focuses.  The first is research on Hidden Markov Models (HMMs) used in medical genomics which result in intractable inference when dependency is introduced between different genome sequences.  The potential impact of this work is in tailored medicine, where drugs could be matched to patients by identifying their closest genetic relative from a set of past patients known to have responded well to particular treatments.


Also related to genetics and biomedical applications, I have developed a strong interest in research into privacy preserving techniques whereby patients can donate potentially sensitive information with mathematically provable guarantees of anonymity and secrecy of their personal data, whilst still allowing researchers to perform statistical analyses with the data.


The third area relates to a core tool used in Bayesian statistical inference, Markov chain Monte Carlo (MCMC).  Algorithmically MCMC is inherently sequential in nature which is discordant with current trends in computing, where massively parallel architectures are increasingly dominant. This is leading to a situation where growth in model complexity is exceeding advances in the serial execution speed of computers.  As such, I have a particular interest in high performance computing architectures such as GPUs and the development of statistical methodology which is amenable to implementation in such environments.


Finally, I maintain an interest in reliability theory which formed part of my PhD work


Links


Research group homepage


Personal academic homepage


 

Network Design