Name
Abstract | ||
|---|---|---|
We address the problem of encoding a graph of order n into a graph of order k <; n in a way to minimize reconstruction error. We characterize this encoding in terms of a particular factorization of the adjacency matrix of the original graph. The factorization is determined as the solution of a discrete optimization problem, which is for convenience relaxed into a continuous, but equivalent, one. We propose a new multiplicative update rule for the optimization task. Experiments are conducted to assess the effectiveness of the proposed approach. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/ICPR.2014.23 | Pattern Recognition |
Keywords | Field | DocType |
graph theory,matrix decomposition,optimisation,adjacency matrix,discrete optimization problem,graph compression,graph encoding,matrix factorization approach,multiplicative update rule | Adjacency matrix,Adjacency list,Factor graph,Discrete mathematics,Strength of a graph,Spectral graph theory,Graph energy,Computer science,Graph bandwidth,Graph partition | Conference |
Volume | ISSN | Citations |
9370 | 1051-4651 | 3 |
PageRank | References | Authors |
0.38 | 11 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Farshad Nourbakhsh | 1 | 13 | 3.02 |
| Samuel Rota Bulo | 2 | 161 | 12.05 |
| Marcello Pelillo | 3 | 1888 | 150.33 |