Tae Yeon Kim

Communication-Efficient Semantic Split Federated Learning via Variational Information Bottleneck Driven Dynamic Slicing

This letter proposes a novel channel-aware semantic SFL (CA-SSFL) framework. The client model employs a dual-masking architecture, where a distributed variational information bottleneck (VIB) generates a semantic mask by prioritizing feature dimensions, while a signal-to-noise ratio (SNR)-driven module produces a dynam...