Image Classification on CIFAR-10 ρ=0.4 (test)
78.07AccuracyHARMONY
Evaluation results
| Method | Links | ||
|---|---|---|---|
| HARMONYBackbone=VGG-11, Number of Clients (N)=50, Data Partitioning (Shard)=22026.05 | 78.07 | 48.23 | |
| SplitGPBackbone=VGG-11, Number of Clients (N)=50, Data Partitioning (Shard)=2, Personalization Intensity (λ)=0.22026.05 | 75.13 | 32.66 | |
| SplitGP + FTBackbone=VGG-11, Number of Clients (N)=50, Data Partitioning (Shard)=2, Personalization Intensity (λ)=0.2, Evaluation Protocol (Fine-tuning)=on-device fine-tuning2026.05 | 73.66 | 50.73 | |
| SplitGPBackbone=VGG-11, Number of Clients (N)=50, Data Partitioning (Shard)=2, Personalization Intensity (λ)=0.92026.05 | 67.69 | 52.99 | |
| Multi-exit SFLBackbone=VGG-11, Number of Clients (N)=50, Data Partitioning (Shard)=22026.05 | 63.06 | 54.92 | |
| SplitFedBackbone=VGG-11, Number of Clients (N)=50, Data Partitioning (Shard)=22026.05 | 46.04 | 146.72 |