Image Classification on CIFAR-100 Practical heterogeneous (alpha = 0.1, K = 20) (test)
54.01AccuracyCAFedCL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CAFedCLBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 54.01 | 3.65 | |
| FedProtoBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 52.81 | 6.58 | |
| FedTGPBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 51.3 | 8.59 | |
| FedRCLBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 49.99 | 8.29 | |
| MP-FedCLBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 49.7 | 6.26 | |
| MOONBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 46.23 | 5.15 | |
| FedLCBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 44.75 | 4.7 | |
| FedProxBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 31.42 | 4.25 | |
| FedAvgBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 29.74 | 4.08 | |
| FedProcBackbone=ResNet-18, Normalization=Group Normalization, Local Training Epochs=5, Optimizer=SGD, Number of runs=52026.03 | 18.98 | 5.62 |