Name
Abstract | ||
|---|---|---|
The construction of a meaningful graph plays a crucial role in the emerging field of signal processing on graphs. In this paper, we address the problem of learning graph Laplacians, which is similar to learning graph topologies, such that the input data form graph signals with smooth variations on the resulting topology. We adopt a factor analysis model for the graph signals and impose a Gaussian probabilistic prior on the latent variables that control these graph signals. We show that the Gaussian prior leads to an efficient representation that favours the smoothness property of the graph signals, and propose an algorithm for learning graphs that enforce such property. Experiments demonstrate that the proposed framework can efficiently infer meaningful graph topologies from only the signal observations. |
Year | Venue | Keywords |
|---|---|---|
2015 | 2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP) | Graph learning, graph signal processing, representation theory, factor analysis, Gaussian prior |
Field | DocType | ISSN |
Graph property,Computer science,Theoretical computer science,Artificial intelligence,Voltage graph,Discrete mathematics,Line graph,Pattern recognition,Null graph,Butterfly graph,Lattice graph,Graph (abstract data type),Complement graph | Conference | 1520-6149 |
Citations | PageRank | References |
12 | 0.63 | 7 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xiaowen Dong | 1 | 249 | 22.07 |
| Dorina Thanou | 2 | 17 | 1.38 |
| Pascal Frossard | 3 | 3015 | 230.41 |
| Pierre Vandergheynst | 4 | 3576 | 208.25 |