Gatsby Computational Neuroscience Unit
17 Queen Square
London - WC1N 3AR
I am a lecturer with the Gatsby Computational Neuroscience Unit, part of the Centre for Computational Statistics and Machine Learning at UCL.
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.
Three variable interaction tests and software
Hilbert Space Embeddings of Predictive State Representations, UAI 2013
Smooth Operators , ICML 2013.
Optimal kernel choice for large-scale two-sample tests , NIPS 2012. Code is available.
EJS paper, On the empirical estimation of integral probability metrics
NIPS 2012 workshop on the Confluence between Kernel Methods and Graphical Models (with Le Song and Alex Smola)
NIPS 2012 workshop on Modern Nonparametric Methods in Machine Learning (with Sivaraman Balakrishnan, Mladen Kolar, John Lafferty, Han Liu, Tong Zhang)
arXiv tech report on the relation between energy distance/brownian distance covariance and kernel statistics on probabilities (MMD/HSIC)