Description Usage Arguments Value References Examples
Graph Diffusion Distance (nd.gdd
) quantifies the difference between two weighted graphs of same size. It takes
an idea from heat diffusion process on graphs via graph Laplacian exponential kernel matrices. For a given
adjacency matrix A, the graph Laplacian is defined as
L := D-A
where D_{ii}=∑_j A_{ij}. For two adjacency matrices A_1 and A_2, GDD is defined as
d_{gdd}(A_1,A_2) = max_t √{\| \exp(-tL_1) -\exp(-tL_2) \|_F^2}
where \exp(\cdot) is matrix exponential, \|\cdot\|_F a Frobenius norm, and L_1 and L_2 Laplacian matrices corresponding to A_1 and A_2, respectively.
1 |
A |
a list of length N containing adjacency matrices. |
out.dist |
a logical; |
vect |
a vector of parameters t whose values will be used. |
a named list containing
an (N\times N) matrix or dist
object containing pairwise distance measures.
an (N\times N) matrix whose entries are maximizer of the cost function.
hammond_graph_2013NetworkDistance
1 2 3 4 5 6 7 8 9 10 11 12 |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.