LATIDIA · Ciberseguridad
SplitReAlign: realineación de la representación con reconocimiento de particiones para el desaprendizaje federado dividido
arXiv:2611.25301v1 Tipo de anuncio: cruzado Resumen: El aprendizaje federado dividido (SFL) distribuye las redes neuronales entre los clientes y un servidor, acoplando el conocimiento específico de cada clase a través de sus representaciones. Este acoplamiento...
WhatsApp ↗Telegram ↗
La noticia
arXiv:2611.25301v1 Announce Type: cross Abstract: Split federated learning (SFL) distributes neural networks across clients and a server, coupling class-specific knowledge across their representations. This coupling complicates split federated unlearning (SFU), where class-wise forgetting must remove target knowledge while preserving retained utility under heterogeneous representation spaces and asymmetric update constraints. We propose SplitReAlign, a partition-aware representation realignment framework for class-wise SFU. It constructs endpoint-specific discriminative subspaces coordinated through a shared semantic anchor, selectively realigning forgotten representations while penalizing changes outside the active subspaces. To accommodate asymmetric update constraints, SplitReAlign leverages gradients through frozen servers for Client-only updates and fixed client-feature statistics for Server-only guidance, while Joint updates optimize both partitions. Experiments