HPFL-CN: Communication-Efficient Hierarchical Personalized Federated Edge Learning via Complex Network Feature Clustering

Self-Evolving Integrated Vertical Heterogeneous Networks

“…Hence, a personalized variant of FL seems to be a better solution. This personalization keeps all the benefits of the federated learning architecture, and, by structure, leads to a more personalized model for each user across vertical tiers [214]. By employing personalized FL, each vertical tier can be enabled to make local decisions on its local network management aspects while transmitting its local ML model to a centralized unit in order to develop and train a global model which is a necessity to ensure full integration and coordination between the tiers.…”