Computes the Laplacian matrix of a graph on the basis of an observed data matrix, where we assume the data to be Gaussian distributed.
Laplacian matrix of a connected graph with Gaussian data
Computes the Laplacian matrix of a graph on the basis of an observed data matrix, where we assume the data to be Gaussian distributed.
learn_connected_graph(
S,
w0 = "naive",
d = 1,
rho = 1,
maxiter = 10000,
reltol = 1e-05,
verbose = TRUE
)A list containing possibly the following elements:
laplacianestimated Laplacian matrix
adjacencyestimated adjacency matrix
thetaestimated Laplacian matrix slack variable
maxiternumber of iterations taken to reach convergence
convergenceboolean flag to indicate whether or not the optimization converged
a p x p covariance matrix, where p is the number of nodes in the graph
initial vector of graph weights. Either a vector of length p(p-1)/2 or a string indicating the method to compute an initial value.
the nodes' degrees. Either a vector or a single value.
constraint relaxation hyperparameter.
maximum number of iterations.
relative tolerance as a convergence criteria.
whether or not to show a progress bar during the iterations.