Statistical Analysis of Pairwise Connectivity

Analysing correlations between streams of events is an important problem. It arises for example in Neurosciences, when the connectivity neurons should be inferred from spike trains that record neurons’ individual spiking activity. While recently some approaches inferring delayed synaptic connections have been proposed, they are limited types connectivities and delays able to handle, or require computation-intensive procedures. This paper proposes a faster more flexible approach analysing such correlated activity: statistical Analysis Connectivity Events (ACE), based on idea hypothesis testing. first computes any pair source target neuron inter-spike subsequent source- target-spikes. Then, it derives null model distribution uncorrelated neurons. Finally, compares observed this infers pairwise Pearson’s \(\chi ^2\) test statistic. Thus, ACE capable detect with priori unknown, non-discrete (and potentially large) delays, which might vary pairs Since works incrementally, has potential being used online processing. In experimental evaluation, performs comparable better than four baseline approaches, terms AUPRC (reported here), F1, AUROC our website), majority 11 evaluated scenarios.