Vine copula classifiers for the mind reading problem

In this paper we introduce vine copulas to model probabilistic dependencies in supervised classification problems. Vine copulas allow the representation of the dependence structure of multidimensional distributions as a factorization of bivariate pair-copulas. The flexibility of this model lies in the fact that we can mix different types of pair-copulas in a factorization, which allows covering a wide range of types of dependencies, i.e., from independence to much more complex forms of bivariate correlations. This property motivates us to use vine copulas as classifiers, particularly for problems for which the type and strength of bivariate interactions between the variables show a great variability. This is the case of brain signal classification problems where information is represented as multiple time series, each one recorded from different brain region. Our experimental results on a real-word Mind Reading Problem reveal that vine copula-based classifiers perform competitively compared to the four best classification methods presented at the Mind Reading Challenge Competition 2011.

[1]  Jack L. Gallant,et al.  Encoding and decoding in fMRI , 2011, NeuroImage.

[2]  Samuel Kaski,et al.  ICANN/PASCAL2 Challenge: MEG Mind Reading — Overview and Results , 2012 .

[3]  M. Sklar.  Fonctions de repartition a n dimensions et leurs marges , 1959 .

[4]  Samuel Kaski,et al.  Multi-class Gaussian process classification of single-trial MEG based on frequency specific latent features extracted with linear binary classifiers , 2011 .

[5]  Arturo Hernández Aguirre,et al.  Estimation of distribution algorithms based on copula functions , 2011, GECCO '11.

[6]  Roger M. Cooke,et al.  Probability Density Decomposition for Conditionally Dependent Random Variables Modeled by Vines , 2001, Annals of Mathematics and Artificial Intelligence.

[7]  H. Joe.  Asymptotic efficiency of the two-stage estimation method for copula-based models , 2005 .

[8]  H. Joe,et al.  The Estimation Method of Inference Functions for Margins for Multivariate Models , 1996 .

[9]  G. Pfurtscheller,et al.  Optimal spatial filtering of single trial EEG during imagined hand movement. , 2000, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[10]  R. Nelsen.  An Introduction to Copulas , 1998 .

[11]  Concha Bielza,et al.  Regularized logistic regression and multiobjective variable selection for classifying MEG data , 2012, Biological Cybernetics.

[12]  K. Aas,et al.  Models for construction of multivariate dependence – a comparison study , 2009 .

[13]  Delbert Dueck,et al.  Clustering by Passing Messages Between Data Points , 2007, Science.

[14]  T. Bedford,et al.  Vines: A new graphical model for dependent random variables , 2002 .

[15]  Paolo Avesani,et al.  MEG decoding across subjects , 2014, 2014 International Workshop on Pattern Recognition in Neuroimaging.

[16]  Hans-Jochen Heinze,et al.  Predicting the recognition of natural scenes from single trial MEG recordings of brain activity , 2000, NeuroImage.

[17]  G. Pfurtscheller,et al.  Brain-Computer Interfaces for Communication and Control. , 2011, Communications of the ACM.

[18]  H. Shimodaira,et al.  Improving predictive inference under covariate shift by weighting the log-likelihood function , 2000 .

[19]  Gal Elidan,et al.  Copula Bayesian Networks , 2010, NIPS.

[20]  H. Joe.  Families of $m$-variate distributions with given margins and $m(m-1)/2$ bivariate dependence parameters , 1996 .

[21]  Heikki Huttunen,et al.  Mind reading with regularized multinomial logistic regression , 2013, Machine Vision and Applications.

[22]  A. Frigessi,et al.  Pair-copula constructions of multiple dependence , 2009 .

[23]  M Congedo,et al.  A review of classification algorithms for EEG-based brain–computer interfaces , 2007, Journal of neural engineering.

[24]  G Pfurtscheller,et al.  Using time-dependent neural networks for EEG classification. , 2000, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[25]  T. Forshaw.  Everything you always wanted to know , 1977 .

[26]  Claudia Czado,et al.  Selecting and estimating regular vine copulae and application to financial returns , 2012, Comput. Stat. Data Anal..

[27]  C.W. Anderson,et al.  Comparison of linear, nonlinear, and feature selection methods for EEG signal classification , 2003, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[28]  Klaus-Robert Müller,et al.  Classifying Single Trial EEG: Towards Brain Computer Interfacing , 2001, NIPS.

[29]  Claudia Czado,et al.  Pair-Copula Constructions of Multivariate Copulas , 2010 .

[30]  Miguel A. L. Nicolelis,et al.  Brain–machine interfaces: past, present and future , 2006, Trends in Neurosciences.

[31]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[32]  Alain Rakotomamonjy,et al.  BCI Competition III: Dataset II- Ensemble of SVMs for BCI P300 Speller , 2008, IEEE Transactions on Biomedical Engineering.

[33]  C. Genest,et al.  Everything You Always Wanted to Know about Copula Modeling but Were Afraid to Ask , 2007 .

[34]  B. Rémillard,et al.  Validity of the parametric bootstrap for goodness-of-fit testing in semiparametric models , 2005 .

[35]  Arturo Hernández Aguirre,et al.  Supervised Probabilistic Classification Based on Gaussian Copulas , 2010, MICAI.

[36]  Daniel Berg.  Copula goodness-of-fit testing: an overview and power comparison , 2009 .

[37]  Daniel J. Rosenkrantz,et al.  An analysis of several heuristics for the traveling salesman problem , 2013, Fundamental Problems in Computing.