A hierarchical federated transfer learning (HFTL) method is proposed for digital twin-based vehicular ad hoc networks (DT-VANET). It clusters vehicles by type and adapts models to data heterogeneity and sparsity, improving global model prediction accuracy. Algorithms for cloud server model updates and intra-cluster transfer learning are developed.
The news is relevant to modeling and management of large systems as it proposes an approach to improve prediction accuracy in distributed vehicular networks with data privacy constraints.
Translation and summary are machine-generated from the source. The full article is not republished; its rights belong to the source.
