Research map / Architectures and efficiency / Neural architecture search
Exploration and Estimation for Model Compression
Yanfu Zhang, Shangqian Gao, Heng Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add exploration step (K=1, m=1) (ablation) | VGG-16 pruning · CIFAR-10 | pruned top-1 accuracy: 92.88% -> 93.17% | Improved |
| add exploration step (K=3, m=3) (ablation) | VGG-16 pruning · CIFAR-10 | pruned top-1 accuracy: 92.88% -> 93.33% | Improved |
| add exploration step (K=5, m=5) (ablation) | VGG-16 pruning · CIFAR-10 | pruned top-1 accuracy: 92.88% -> 93.63% | Improved |
| add exploration step (K=1, m=1) (ablation) | ResNet-56 pruning · CIFAR-10 | pruned top-1 accuracy: 93.31% -> 93.46% | Improved |
| add exploration step (K=3, m=3) (ablation) | ResNet-56 pruning · CIFAR-10 | pruned top-1 accuracy: 93.31% -> 93.59% | Improved |
| add exploration step (K=5, m=5) (ablation) | ResNet-56 pruning · CIFAR-10 | pruned top-1 accuracy: 93.31% -> 93.68% | Improved |
| use proposed method instead of IE | ResNet-50 pruning · ImageNet | pruned top-1 accuracy: -1.68% -> -0.10% | Improved |
| use proposed method instead of SCP | ResNet-50 pruning · ImageNet | pruned top-1 accuracy: 0.13% higher in top-1 delta | Improved |
| use proposed method instead of FPGM | ResNet-18 pruning · ImageNet | pruned top-1 accuracy: -1.87% -> -2.01% | Hurt |
| use proposed method instead of IE | VGG-11bn pruning · ImageNet | pruned top-1 accuracy: -0.20% -> +0.31% | Improved |
| harder budget (35% vs 80% FLOPs) makes search slower | proposed pruning search · CIFAR-10 | pruned top-1 accuracy: 50 epochs vs 300 epochs | Mixed |
Closest papers by meaning
- An Information Theory-Inspired Strategy for Automated Network Pruning (IJCV 2025 · similarity 0.90)
- Discrimination-aware Network Pruning for Deep Model Compression (TPAMI 2021 · similarity 0.90)
- Discrete Model Compression With Resource Constraint for Deep Neural Networks (CVPR 2020 · similarity 0.89)
- CHEX: CHannel EXploration for CNN Model Compression (CVPR 2022 · similarity 0.89)
- Network Pruning via Performance Maximization (CVPR 2021 · similarity 0.89)
- Multi-Dimensional Pruning: A Unified Framework for Model Compression (CVPR 2020 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
