Home

My Group

Publications

Teaching

Software

Interesting Things

Dr Gavin Brown University of Manchester School of Computer Science Kilburn Building, Oxford Road Manchester, M13 9PL 0161 275 6190 Who am I? What do I do? I am a member of the Machine Learning and Optimization Group. My research interests can be summarised as: feature selection/extraction with information theoretic methods, Markov Blanket algorithms, ensemble learning (aka multiple classifier systems), and online learning. All of the above in application to two domains: Systems Biology and adaptive compiler optimisation. Or, in less technical jargon:... click here

NEWS: (Jan 2012) The BBC's flagship technology programme "Click" featured our project. You can hear the podcast here. NEWS: (Dec 2011) Paper accepted to AISTATS 'Informative Priors for Markov Blanket Discovery' NEWS: (Nov 2011) Paper accepted to JMLR on Conditional Likelihood for Feature Selection NEWS: (Oct 2011) We are pleased to announce the completion of the REUNITE project. NEWS: (Oct 2011) New grant on multi-core machine learning NEWS: (June 2011) Paper accepted to UAI "Boosting as a Product of Experts" Recent Activities: Talks on the JMLR paper I have been touring somewhat, giving talks about our recent JMLR paper - thanks for the invites everyone! Visiting... Surrey Elec Eng, Birmingham Computer Science, Manchester Medical School.... next scheduled talk: Oxford (Mathematics Dept) in May. REUNITE project featured by BBC World Service The BBC's flagship technology programme "Click" recently featured our project. You can hear the podcast here. AstraZeneca MSc Research Bursaries I am currently investigating biomarkers for lung cancer analysis with AstraZeneca Research. AZ have sponsored our students this year, under their predictive safety science initiative. Invited Doctoral Lecture Course University of Cagliari, Sardinia I delivered a series of 8 invited lectures in Cagliari - see the course webpages here. Invited lecture at IEEE symposium I am delivering a keynote at the 2011 IEEE symposium on Computational Intelligence, on the topic of computational intelligence in dynamic and uncertain environments. New book chapter - Ensemble Learning I wrote an invited book chapter for the Springer Encyclopedia of Machine Learning. You can see also the typeset article here. "The study of ensemble methods, with model outputs considered for their abstract properties rather than the specifics of the algorithm which produced them, allows for a wide impact across many fields of study. If we can understand precisely why, when, and how particular ensemble methods can be applied successfully, we would have made progress toward a powerful new tool for Machine Learning: the ability to automatically exploit the strengths and weaknesses of different learning systems." New PhD (Dec 2010) : - Manuela Zanda completed her PhD, entitled ``A Probabilistic Perspective on Ensemble Diversity''. A copy of her thesis can be downloaded here. New Grant (9th Sept 2010) - EPSRC KTA, Reuniting Refugees with Computational Intelligence. REUNITE is a research project aiming to utilise crowdsourcing and machine learning techniques to help reunite those separated by conflict and natural disaster. Imagine the following scenario. A disaster occurs in a remote part of the developing world. The local population are forced to flee their homes. Many are separated from their family and friends. With no mobile or Internet communication, finding loved ones in the aftermath of a disaster is incredibly difficult. Relief organisations go to great lengths to help people find those they are missing. The system we are developing aims to make this process easier, faster and more secure. Article in THE (3 June 2010): I had an article about computer science education in the Times Higher this week. Are you looking for the Computing at School group? Or for the Manchester Schools' Animation Competition? The Animation competition is an effort led by Toby Howard, to encourage schoolchildren to learn the concepts of computational thinking, and I strongly encourage all to take note! Invited plenary talk at MCS 2010 I gave an invited talk at the Intl Workshop on Multiple Classifier Systems 2010, entitled Some Thoughts at the Interface of Ensemble Methods and Feature Selection. It was repeated with (slightly) adapted slides for Microsoft Research Cairo, New PhD (Nov 2009) : - Amir Ahmad completed his PhD, entitled ``Data Transformations for Decision Tree Ensembles''. A copy of his thesis can be downloaded here. AISTATS 2009 paper - Feature Selection with Information Theory The traditional approach to so-called filter methods in feature selection is to construct a criterion to measure the utility of any given feature. The more sophisticated methods penalize feature-feature correlations (`redundancy') with various penalty terms. The last 15 years have produced a flood of papers advocating different penalty terms. My recent work shows that the vast majority of these can be naturally derived from a single framework, using multivariate information theory. The work reveals that there exists a natural, smooth space space of feature selection criteria, where each paper over the last 15 years corresponds to one point. Most of the space has never been explored. See the AISTATS 2009 paper for details. Invited plenary at UK-KDD 2009 - Feature Selection by Filters, a Unifying Perspective I gave an invited talk at UK-KDD 2009. New Grant - Dynamic Ensemble Techniques (EPSRC grant EP/F023855/1) With colleagues at Bristol, I am investigating how dynamic ensemble techniques can tackle multi-step (control) and nonstationary problems. This is in collaboration with Tim Kovacs, James Marshall and Jeremy Wyatt, conducted under our EPSRC funded ADEPT project. New Grant - Machine Learning for Multi-Core Computers (EPSRC grant EP/G000662/1) The computer industry is undergoing the "multi-core" revolution. When you buy a PC off the shelf these days, it is inevitably "dual-core" or "quad-core". This idea of more and more CPU "cores" executing in parallel is expected to continue to the hundreds and thousands. The problem of coordinating these cores is challenging and unsolved. With Mikel Lujan and Jeremy Singer I am working on applying Machine Learning to this problem, conducted under our EPSRC funded iTLS project. IEEE TNN paper on Sparse Distributed Memories In a project with Steve Furber I found that sparse distributed memory models like the correlation matrix memories of Wilshaw and Kanerva could give significant insights into the design of fault tolerant computer architectures. This resulted in a IEEE TNN paper available here. Ensemble Learning I worked for a long while on the issue of diversity in ensembles, with Jeremy Wyatt. A summary of the work can be found on this page. A slightly less optimistic (but rather insightful) take on the field is found here. Image Feature Extraction I did a nice project with Honda several years ago, which turned into a patent, on image feature extraction - I follow up little avenues on this occasionally. Throughout this time I have maintained an interest in evolutionary speciation and optimisation, which has spun off into several useful collaborations.