Name
Abstract | ||
|---|---|---|
Recently, Kishida et al. have proposed the deep learning based parameter (weighting factors) optimization method for the average consensus problem with a complex network, which can significantly accelerate its convergence performance. However, the optimized parameters cannot be applied into the case with different network topology from the one used in the training. This study tries to make the optimized parameters versatile for the network topology. Specifically, we propose a modified deep learning aided optimization method considering the stochastic feature of network which can be captured by the network centrality. The training in the proposed method is done with the restriction caused by the centrality. This work considers two types of centralities, i.e., degree and eigenvector centralities. We further propose the leaning method considering the relationship of centralities at the connected nodes. Simulation results show that the convergence performance of the proposed system can approach to the conventional method even with the different network from the ones used in the training. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/VTC2021-FALL52928.2021.9625549 | 2021 IEEE 94TH VEHICULAR TECHNOLOGY CONFERENCE (VTC2021-FALL) |
Keywords | DocType | ISSN |
Consensus problem, data-driven optimization, network centrality | Conference | 2577-2465 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Shoya Ogawa | 1 | 0 | 0.34 |
| Koji Ishii | 2 | 0 | 0.34 |