RingSFL:An Adaptive Split Federated Learning Towards Taming Client Heterogeneity

Keyphrases

• Data Privacy 100%
• Federated Learning 100%
• Client Heterogeneity 100%
• Split Federated Learning 100%
• Identically Distributed 66%
• Training Effectiveness 66%
• Eavesdropper 66%
• Training Data 33%
• Benchmarking Method 33%
• Complete Model 33%
• Convergence Performance 33%
• Local Model 33%
• Learning Scheme 33%
• Ring Topology 33%
• Distributed Learning 33%
• Privacy Leakage 33%
• Heterogeneous Clients 33%
• Propagation Length 33%
• Client Data 33%
• Vanilla 33%
• Distributed Datasets 33%
• Stragglers 33%
• Straggler Effect 33%
• Split Mechanism 33%

Computer Science

• Federated Learning 100%
• Data Privacy 75%
• Privacy Leakage 25%
• Distributed Learning 25%
• Training Data 25%
• Learning Scheme 25%
• Experimental Result 25%
• Convergence Performance 25%