Arthur Gretton

Arthur Gretton

Auto-generated profile, based on publicly available information

Arthur Gretton

Gatsby Computational Neuroscience Unit University College London

Research Focus & Metrics

Academic Career Length

25Years (2002-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 July 2006 in Bioinformatics
Verified
Journal Article
Open access
KBKarsten M. Borgwardt
Karsten M. Borgwardt
AGArthur Gretton
Arthur Gretton
MRMalte J. Rasch
Malte J. Rasch
HKHans-Peter Kriegel
Hans-Peter Kriegel
BSBernhard Schölkopf
Bernhard Schölkopf
ASAlex J. Smola
Alex J. Smola
Motivation: Many problems in data integration in bioinformatics can be posed as one common question: Are two sets of observations generated by the same distribution? We propose a kernel-based statistical test for this problem, based on the fact that two distributions are different if and only if there exists at least one function having different expectation on the two distributions. Consequently we use the maximum discrepancy between function means as the basis of a test statistic. The Maximum Mean Discrepancy (MMD) can take advantage of the kernel trick, which allows us to apply it not only to vectors, but strings, sequences, graphs, and other common structured data types arising in molecular biology. Results: We study the practical feasibility of an MMD-based test on three central data integration tasks: Testing cross-platform comparability of microarray data, cancer diagnosis, and data-content based schema matching for two different protein function classification schemas. In all of these experiments, including high-dimensional ones, MMD is very accurate in finding samples that were generated from the same distribution, and outperforms its best competitors. Conclusions: We have defined a novel statistical test of whether two samples are from the same distribution, compatible with both multivariate and structured data, that is fast, easy to implement, and works well, as confirmed by our experiments. Availability: Contact: kb@dbs.ifi.lmu.de

1 January 2005
Verified
Book Chapter
Closed access
PDF file
AGArthur Gretton
Arthur Gretton
OBOlivier Bousquet
Olivier Bousquet
ASAlex Smola
Alex Smola
BSBernhard Schölkopf
Bernhard Schölkopf

1 October 2013 in The Annals of Statistics
Verified
Journal Article
Open access
PDF file
DSDino Sejdinovic
Dino Sejdinovic
BSBharath Sriperumbudur
Bharath Sriperumbudur
AGArthur Gretton
Arthur Gretton
KFKenji Fukumizu
Kenji Fukumizu
We provide a unifying framework linking two classes of statistics used in two-sample and independence testing: on the one hand, the energy distances and distance covariances from the statistics literature; on the other, maximum mean discrepancies (MMD), that is, distances between embeddings of distributions to reproducing kernel Hilbert spaces (RKHS), as established in machine learning. In the case where the energy distance is computed with a semimetric of negative type, a positive definite kernel, termed distance kernel, may be defined such that the MMD corresponds exactly to the energy distance. Conversely, for any positive definite kernel, we can interpret the MMD as energy distance with respect to some negative-type semimetric. This equivalence readily extends to distance covariance using kernels on the product space. We determine the class of probability distributions for which the test statistics are consistent against all alternatives. Finally, we investigate the performance of the family of distance kernels in two-sample and independence tests: we show in particular that the energy distance most commonly employed in statistics is just one member of a parametric family of kernels, and that other choices from this family can yield more powerful tests.

28 May 2008 in Journal of Neuroscience
Verified
Journal Article
Open access
PDF file
ABAndrei Belitski
Andrei Belitski
AGArthur Gretton
Arthur Gretton
CMCesare Magri
Cesare Magri
YMYusuke Murayama
Yusuke Murayama
MMMarcelo A. Montemurro
Marcelo A. Montemurro
Stefano PanzeriStefano Panzeri
Stefano Panzeri
Local field potentials (LFPs) reflect subthreshold integrative processes that complement spike train measures. However, little is yet known about the differences between how LFPs and spikes encode rich naturalistic sensory stimuli. We addressed this question by recording LFPs and spikes from the primary visual cortex of anesthetized macaques while presenting a color movie. We then determined how the power of LFPs and spikes at different frequencies represents the visual features in the movie. We found that the most informative LFP frequency ranges were 1–8 and 60–100 Hz. LFPs in the range of 12–40 Hz carried little information about the stimulus, and may primarily reflect neuromodulatory inputs. Spike power was informative only at frequencies <12 Hz. We further quantified “signal correlations” (correlations in the trial-averaged power response to different stimuli) and “noise correlations” (trial-by-trial correlations in the fluctuations around the average) of LFPs and spikes recorded from the same electrode. We found positive signal correlation between high-gamma LFPs (60–100 Hz) and spikes, as well as strong positive signal correlation within high-gamma LFPs, suggesting that high-gamma LFPs and spikes are generated within the same network. LFPs <24 Hz shared strong positive noise correlations, indicating that they are influenced by a common source, such as a diffuse neuromodulatory input. LFPs 40 Hz and with spikes, suggesting that low-frequency LFPs reflect neural processes that in natural conditions are fully decoupled from those giving rise to spikes and to high-gamma LFPs.

1 January 2007
Verified
Book Chapter
Closed access
PDF file
ASAlex Smola
Alex Smola
AGArthur Gretton
Arthur Gretton
LSLe Song
Le Song
BSBernhard Schölkopf
Bernhard Schölkopf

1 January 2012 in Electronic Journal of Statistics
Verified
Journal Article
Open access
PDF file
BSBharath K. Sriperumbudur
Bharath K. Sriperumbudur
KFKenji Fukumizu
Kenji Fukumizu
AGArthur Gretton
Arthur Gretton
BSBernhard Schölkopf
Bernhard Schölkopf
GLGert R. G. Lanckriet
Gert R. G. Lanckriet
Given two probability measures, $\mathbb{P}$ and $\mathbb{Q}$ defined on a measurable space, $S$, the integral probability metric (IPM) is defined as $$\gamma_{\EuScript{F}}(\mathbb{P},\mathbb{Q})=\sup\left\{\left\vert \int_{S}f\,d\mathbb{P}-\int_{S}f\,d\mathbb{Q}\right\vert\,:\,f\in\EuScript{F}\right\},$$ where $\EuScript{F}$ is a class of real-valued bounded measurable functions on $S$. By appropriately choosing $\EuScript{F}$, various popular distances between $\mathbb{P}$ and $\mathbb{Q}$, including the Kantorovich metric, Fortet-Mourier metric, dual-bounded Lipschitz distance (also called the Dudley metric), total variation distance, and kernel distance, can be obtained. In this paper, we consider the problem of estimating $\gamma_{\EuScript{F}}$ from finite random samples drawn i.i.d. from $\mathbb{P}$ and $\mathbb{Q}$. Although the above mentioned distances cannot be computed in closed form for every $\mathbb{P}$ and $\mathbb{Q}$, we show their empirical estimators to be easily computable, and strongly consistent (except for the total-variation distance). We further analyze their rates of convergence. Based on these results, we discuss the advantages of certain choices of $\EuScript{F}$ (and therefore the corresponding IPMs) over others—in particular, the kernel distance is shown to have three favorable properties compared with the other mentioned distances: it is computationally cheaper, the empirical estimate converges at a faster rate to the population value, and the rate of convergence is independent of the dimension $d$ of the space (for $S=\mathbb{R}^{d}$). We also provide a novel interpretation of IPMs and their empirical estimators by relating them to the problem of binary classification: while the IPM between class-conditional distributions is the negative of the optimal risk associated with a binary classifier, the smoothness of an appropriate binary classifier (e.g., support vector machine, Lipschitz classifier, etc.) is inversely related to the empirical estimator of the IPM between these class-conditional distributions.