Image Classification on CIFAR-10 (test) (Accuracy and Throughput)
96.41Accuracy (CIFAR-10 Test)SGD
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SGDBackbone=WRN-28-10, Sparsity=-2026.03 | 96.41 | 752.3 | |
| GSAMBackbone=WRN-28-10, Sparsity=90%2026.03 | 95.4 | 340.19 | |
| SAMBackbone=WRN-28-10, Sparsity=90%2026.03 | 95.37 | 354.94 | |
| ESAMBackbone=WRN-28-10, Sparsity=90%2026.03 | 95.31 | 305.76 | |
| ZO-SAMBackbone=WRN-28-10, Sparsity=90%2026.03 | 95.02 | 576.01 | |
| LS(k=5)Backbone=WRN-28-10, Sparsity=90%2026.03 | 94.89 | 542.63 | |
| SGDBackbone=ResNet-32, Sparsity=-2026.03 | 94.58 | 5,673.95 | |
| LS(k=10)Backbone=WRN-28-10, Sparsity=90%2026.03 | 94.17 | 593.02 | |
| SAMBackbone=ResNet-32, Sparsity=90%2026.03 | 93.77 | 2,704.84 | |
| GSAMBackbone=ResNet-32, Sparsity=90%2026.03 | 93.72 | 2,701 | |
| ESAMBackbone=ResNet-32, Sparsity=90%2026.03 | 93.67 | 2,297.23 | |
| ZO-SAMBackbone=ResNet-32, Sparsity=90%2026.03 | 93.5 | 4,349.53 | |
| LS(k=5)Backbone=ResNet-32, Sparsity=90%2026.03 | 93.42 | 3,980.62 | |
| LS(k=10)Backbone=ResNet-32, Sparsity=90%2026.03 | 93.03 | 4,272.22 |