Image Classification on CIFAR-10 (test) (Beta-Weighted Accuracy)
79.7Accuracy (beta=0.1)OnDev-LCT-8/1
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| OnDev-LCT-8/1Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 79.7 | 84.74 | 86.77 | |
| OnDev-LCT-4/1Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 78.1 | 84.01 | 86.12 | |
| OnDev-LCT-2/1Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 76.85 | 82.81 | 84.95 | |
| OnDev-LCT-1/1Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 76.02 | 82.25 | 84.24 | |
| ResNet-44Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 72.56 | 77.73 | 80.55 | |
| ResNet-32Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 71.55 | 78.05 | 80.39 | |
| ResNet-20Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 67.55 | 75.43 | 77.72 | |
| CCT-2/2Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 66.45 | 74.86 | 78.42 | |
| CCT-4/2Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 66.38 | 75.36 | 78.55 | |
| MobileNetv2/0.5Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 60.5 | 72.35 | 74.27 | |
| MobileNetv2/0.2Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 56.55 | 67.07 | 69.44 | |
| ViT-Lite-2/8Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 49.03 | 58.57 | 60.02 | |
| ViT-Lite-1/8Federated Learning algorithm=FedAvg, Number of clients=10, Evaluation protocol=Best of 3 runs2024.01 | 48.53 | 58.05 | 59.7 |