Gatsby Computational Neuroscience Unit
17 Queen Square
London - WC1N 3AR
+44 (0)7795 291 705
I am a lecturer with the Gatsby Computational Neuroscience Unit, part of the Centre for Computational Statistics and Machine Learning at UCL. A short biography.
My current research focus is on using kernel methods to reveal properties and relations in data. A first application is in measuring distances between probability distributions. These distances can be used to determine strength of dependence, for example in measuring how strongly two bodies of text in different languages are related; testing for similarities in two datasets, which can be used in attribute matching for databases (that is, automatically finding which fields of two databases correspond); and testing for conditional dependence, which is useful in detecting redundant variables that carry no additional predictive information, given the variables already observed. I am also working on applications of kernel methods to inference in graphical models, where the relations between variables are learned directly from training data: applications include cross-language document retrieval, depth prediction from still images, and protein configuration prediction.
Course notes from my tutorial at the kernel methods for big data workshop. See the teaching page .
Updates made Feb. 2014 to Kernel Adaptive Metropolis-Hastings , with new experiments on adaptive sampling for pseudo-marginal MCMC. Code.
Paper on independence testing for random processes. Code.
Paper on regressing from probability distributions to real numbers. Code.
Kernel Bayes' Rule, JMLR, December 2013.
Three variable interaction tests and software (NIPS 2013)
Kernel two-sample tests based on low variance, asymptotically Normal kernel statistics, and software (NIPS 2013)
Paper in Annals of Statistics, on the relation between energy distance/brownian distance covariance and kernel statistics on probabilities (MMD/HSIC).