Tampere University of Technology

TUTCRIS Research Portal

Mind reading with regularized multinomial logistic regression

Research output: Scientific - peer-reviewArticle

Details

Original languageEnglish
Pages (from-to)1311-1325
JournalMachine Vision and Applications
Volume24
Issue number6
DOIs
StatePublished - 2013
Publication typeA1 Journal article-refereed

Abstract

In this paper, we consider the problem of multinomial classification of magnetoencephalography (MEG) data. The proposed method participated in the MEG mind reading competition of ICANN'11 conference, where the goal was to train a classifier for predicting the movie the test person was shown. Our approach was the best among 10 submissions, reaching accuracy of 68 % of correct classifications in this five category problem. The method is based on a regularized logistic regression model, whose efficient feature selection is critical for cases with more measurements than samples. Moreover, a special attention is paid to the estimation of the generalization error in order to avoid overfitting to the training data. Here, in addition to describing our competition entry in detail, we report selected additional experiments, which question the usefulness of complex feature extraction procedures and the basic frequency decomposition of MEG signal for this application.

Publication forum classification

Field of science, Statistics Finland

Downloads statistics

No data available