We propose a straggler-aware resource allocation in semi-decentralized federated learning for large-scale machine-learning models over OTNs. Simulations show it achieves only 2.55% gap to MILP while reducing runtime by 99.78%, and outperforms the SoTA heuristic with 7% higher task success rate and 41% less runtime.
Straggler-Aware Resource Allocation in Semi-Decentralized Federated Learning for Large-Scale Models Over OTNs
Zhang, Qiaolun;Musumeci, Francesco;Tornatore, Massimo;
2025-01-01
Abstract
We propose a straggler-aware resource allocation in semi-decentralized federated learning for large-scale machine-learning models over OTNs. Simulations show it achieves only 2.55% gap to MILP while reducing runtime by 99.78%, and outperforms the SoTA heuristic with 7% higher task success rate and 41% less runtime.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
ECOC2025_0502.pdf
Accesso riservato
:
Pre-Print (o Pre-Refereeing)
Dimensione
1.09 MB
Formato
Adobe PDF
|
1.09 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
