Gradient based Information Aggregation of GNN for Precoder Learning

Employing graph neural networks (GNNs) for learning the multiuser multi-input multi-output precoder has gained significant attention recently. By modeling the precoder optimization problem in a graph format, GNN can effectively capture the representation of the precoder by leveraging the information...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE Vehicular Technology Conference s. 1 - 6
Hlavní autoři: Chen, Shiyong, Han, Shengqian, Li, Yang
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 10.10.2023
Témata:
Behavioral sciences
GNN
gradient descent
Graph neural networks
information aggregation
Mathematical models
Numerical models
Optimization
precoder
Simulation
Vehicular and wireless technologies
ISSN:2577-2465
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Employing graph neural networks (GNNs) for learning the multiuser multi-input multi-output precoder has gained significant attention recently. By modeling the precoder optimization problem in a graph format, GNN can effectively capture the representation of the precoder by leveraging the information aggregated and propagated across the graph. In this paper, we strive to design the information aggregation mechanism of GNN. By analyzing the behavior of the numerical gradient descent algorithm for precoder optimization, we identify the relevant information and the appropriate form for aggregation, enabling us to develop new update equations for GNNs. Simulation results demonstrate the advantages of the proposed GNNs in learning and generalization performance.
ISSN:2577-2465
DOI:10.1109/VTC2023-Fall60731.2023.10333802