Phi_Toolbox.zip (3.28 MB)
Information Integration in Large Brain Networks
Version 4 2019-02-09, 22:27
Version 3 2018-12-28, 08:09
Version 2 2018-10-17, 01:27
Version 1 2018-10-16, 23:18
software
posted on 2019-02-09, 22:27 authored by Daniel TokerDaniel TokerThis toolbox provides code for calculating integrated information in large brain networks, given multivariate Gaussian data recorded from those networks. The code can implement the spectral clustering-based algorithm described in Toker and Sommer, "Information Integration in Large Brain Networks," and can also implement the Queyranne algorithm and the Replica Exchange Markov Chain Monte Carlo (REMCMC) algorithm described in Kitazono et al (2018).
Note that phi_G_Gauss_AL.m, which implements an augmented Lagrangian method for calculating integrated information, is borrowed from Kitazono et al. Also note that the Queyranne algorithm is implemented using the Submodular Function Optimization toolbox, which is included as part of this toolbox, and which was downloaded at the following URL:
https://www.mathworks.com/matlabcentral/fileexchange/20504-submodular-function-optimization
Instructions for how to use the toolbox are included in demo.m.
