SuperSFL: Resource-Heterogeneous Federated Split Learning with Weight-Sharing Super-Networks
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911717446909952 |
|---|---|
| author | Asif, Abdullah Al Yu, Sixing Munoz, Juan Pablo Mazaheri, Arya Jannesari, Ali |
| author_facet | Asif, Abdullah Al Yu, Sixing Munoz, Juan Pablo Mazaheri, Arya Jannesari, Ali |
| contents | SplitFed Learning (SFL) combines federated learning and split learning to enable collaborative training across distributed edge devices; however, it faces significant challenges in heterogeneous environments with diverse computational and communication capabilities. This paper proposes \textit{SuperSFL}, a federated split learning framework that leverages a weight-sharing super-network to dynamically generate resource-aware client-specific subnetworks, effectively mitigating device heterogeneity. SuperSFL introduces Three-Phase Gradient Fusion (TPGF), an optimization mechanism that coordinates local updates, server-side computation, and gradient fusion to accelerate convergence. In addition, a fault-tolerant client-side classifier and collaborative client--server aggregation enable uninterrupted training under intermittent communication failures. Experimental results on CIFAR-10 and CIFAR-100 with up to 100 heterogeneous clients show that SuperSFL converges $2$--$5\times$ faster in terms of communication rounds than baseline SFL while achieving higher accuracy, resulting in up to $20\times$ lower total communication cost and $13\times$ shorter training time. SuperSFL also demonstrates improved energy efficiency compared to baseline methods, making it a practical solution for federated learning in heterogeneous edge environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_02092 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SuperSFL: Resource-Heterogeneous Federated Split Learning with Weight-Sharing Super-Networks Asif, Abdullah Al Yu, Sixing Munoz, Juan Pablo Mazaheri, Arya Jannesari, Ali Distributed, Parallel, and Cluster Computing SplitFed Learning (SFL) combines federated learning and split learning to enable collaborative training across distributed edge devices; however, it faces significant challenges in heterogeneous environments with diverse computational and communication capabilities. This paper proposes \textit{SuperSFL}, a federated split learning framework that leverages a weight-sharing super-network to dynamically generate resource-aware client-specific subnetworks, effectively mitigating device heterogeneity. SuperSFL introduces Three-Phase Gradient Fusion (TPGF), an optimization mechanism that coordinates local updates, server-side computation, and gradient fusion to accelerate convergence. In addition, a fault-tolerant client-side classifier and collaborative client--server aggregation enable uninterrupted training under intermittent communication failures. Experimental results on CIFAR-10 and CIFAR-100 with up to 100 heterogeneous clients show that SuperSFL converges $2$--$5\times$ faster in terms of communication rounds than baseline SFL while achieving higher accuracy, resulting in up to $20\times$ lower total communication cost and $13\times$ shorter training time. SuperSFL also demonstrates improved energy efficiency compared to baseline methods, making it a practical solution for federated learning in heterogeneous edge environments. |
| title | SuperSFL: Resource-Heterogeneous Federated Split Learning with Weight-Sharing Super-Networks |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2601.02092 |