LIM_simulator: LIM_simulator

In BenjiUvA/Learning_Ising_Model: LIMA data simulation

Description

Simulates data from the Learning Ising Model

Usage

LIM_simulator(n = 10, nreps = 100, tau, omega, beta,
  X1 = sample(c(-1, 1), n, T), Hebb = T, e = 0.001, lambda = 0.001)

Arguments

n	is the number of nodes in the network
nreps	is the number of iterations
tau	is a vector with a threshold for each node
omega	is a (symmetrical) matrix with the strength of connection between each nodes. Set the the diagonal to zero.
beta	is the dependency parameter in the Ising model
X1	is the initial configuration of the network (e.g. if there are three nodes, X1 can be c(-1,1,-1))
Hebb	do you want to update omega with Hebb's rule?
e	is the learning parameter of Hebb's rule
lambda	is the decay parameter of Hebb's rule

Value

configurations contains all the configurations of the network over all nreps iterations

beta contains a vector of length()==nreps with values used for beta

mean omega contains a vector with the mean of omega at each configuration

omega t=... contains omega at 5 timepoints between the first and the last iteration (equal intervals)

gibbs entropy a vector with gibbs entropy for each iteration

BenjiUvA/Learning_Ising_Model documentation built on June 8, 2019, 3:44 a.m.