Batched split learning: a novel framework for low-latency and energy-efficient split learning with optimized task scheduling

To address the issues of high energy consumption and slow convergence speed in Split learning, we propose a novel framework Batched Split Learning (BSL). We group edge nodes and deploy different device-side models in each batch, then deploy the corresponding server-side model on the edge server. Edg...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Wireless networks Ročník 31; číslo 6; s. 4245 - 4263
Hlavní autoři:	Li, Zhuo, Zheng, Kang
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York Springer US 01.08.2025 Springer Nature B.V
Témata:	Accuracy Algorithms Clustering Communication Communications Engineering Computer Communication Networks Crushing Deep learning Edge computing Electrical Engineering Energy consumption Energy costs Energy efficiency Engineering Heuristic Heuristic methods Internet of Things IT in Business Learning Machine learning Mixed integer Model updating Networks Nodes Nonlinear programming Optimization Original Paper Scheduling Servers Task scheduling Parallel model training Latency optimization Batched split learning Energy consumption optimization
ISSN:	1022-0038, 1572-8196
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Buďte první, kdo okomentuje tento záznam!