Image Classification on MNIST 100 clients, highly heterog. (test)
96Test AccuracyIVON-ADMM@mg
Evaluation results
| Method | Links | ||
|---|---|---|---|
| IVON-ADMM@mgBackbone=MLP, Communication rounds=1002025.06 | 96 | 0.13 | |
| IVON-ADMMBackbone=MLP, Communication rounds=1002025.06 | 96 | 0.13 | |
| FedDynBackbone=MLP, Communication rounds=1002025.06 | 94.9 | 0.18 | |
| FedLap-CovBackbone=MLP, Communication rounds=1002025.06 | 94.9 | 0.18 | |
| FedLapBackbone=MLP, Communication rounds=1002025.06 | 94.4 | 0.22 | |
| IVON-ADMM@mgBackbone=MLP, Communication rounds=502025.06 | 94.2 | 0.19 | |
| IVON-ADMMBackbone=MLP, Communication rounds=502025.06 | 94.2 | 0.19 | |
| FedAvg/FedProxBackbone=MLP, Communication rounds=1002025.06 | 94 | 0.22 | |
| FedLap-CovBackbone=MLP, Communication rounds=502025.06 | 92.9 | 0.29 | |
| FedDynBackbone=MLP, Communication rounds=502025.06 | 92.6 | 0.33 | |
| FedLapBackbone=MLP, Communication rounds=502025.06 | 92 | 0.38 | |
| FedAvg/FedProxBackbone=MLP, Communication rounds=502025.06 | 91.8 | 0.36 | |
| FedLap-CovBackbone=MLP, Communication rounds=252025.06 | 91 | 0.47 | |
| IVON-ADMM@mgBackbone=MLP, Communication rounds=252025.06 | 91 | 0.31 | |
| IVON-ADMMBackbone=MLP, Communication rounds=252025.06 | 91 | 0.31 | |
| FedDynBackbone=MLP, Communication rounds=252025.06 | 89.8 | 0.6 | |
| FedLapBackbone=MLP, Communication rounds=252025.06 | 89.7 | 0.63 | |
| FedAvg/FedProxBackbone=MLP, Communication rounds=252025.06 | 89.6 | 0.59 |