You are here

Danilo Benozzo and Emanuele Olivetti win the 2nd prize of Biomag2014 Causality Competition

News date: 
Monday, 25 August, 2014

The Causality Competition Causal2014 was organized in conjunction with the conference Biomag2014 and it aimed to promote research in the field of inferring causality among MEG time series. The team developed their solution by looking at the problem as a classification task. Thus, after the definition of a proper feature space, a classifier was trained to infer the causality structure within each trial of the competition data set.

This research area has a strong influence on studying the effective connectivity in the brain through the analysis of the magneto/electroencephalography signals. In particular, the goal of the competition was to estimate the direct causal interactions in each trial of a provided data set. The data set contained 1000 trials each one made of three time series generated by a multivariate autoregressive model, according to a predefined causality structure.