Image Classification on CIFAR-10 (test) (Accuracy and FLOPs)
93.65Accuracy (%)PruneFuse
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PruneFuseBudget (b)=50%, Params=0.21 M, Pruning ratio (p)=0.5, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 93.65 | 3.67 | |
| PruneFuseBudget (b)=50%, Params=0.08 M, Pruning ratio (p)=0.7, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 93.4 | 1.32 | |
| PruneFuseBudget (b)=50%, Params=0.03 M, Pruning ratio (p)=0.8, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 93.32 | 0.58 | |
| PruneFuseBudget (b)=50%, Params=0.14 M, Pruning ratio (p)=0.6, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 93.08 | 2.35 | |
| Baseline (AL)Budget (b)=50%, Params=0.85 M, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 93 | 14.66 | |
| PruneFuseBudget (b)=40%, Params=0.21 M, Pruning ratio (p)=0.5, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 92.77 | 2.24 | |
| Baseline (AL)Budget (b)=40%, Params=0.85 M, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 92.24 | 8.94 | |
| PruneFuseBudget (b)=30%, Params=0.21 M, Pruning ratio (p)=0.5, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 91.44 | 1.16 | |
| Baseline (AL)Budget (b)=30%, Params=0.85 M, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 90.85 | 4.64 | |
| PruneFuseBudget (b)=20%, Params=0.21 M, Pruning ratio (p)=0.5, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 88.35 | 0.44 | |
| Baseline (AL)Budget (b)=20%, Params=0.85 M, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 87.74 | 1.76 | |
| PruneFuseBudget (b)=10%, Params=0.21 M, Pruning ratio (p)=0.5, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 80.92 | 0.08 | |
| Baseline (AL)Budget (b)=10%, Params=0.85 M, Backbone=ResNet-56, Sampling Strategy=Least-Confidence2026.03 | 80.53 | 0.31 |