Research map / Architectures and efficiency / Pruning and efficient tuning
Network Pruning via Performance Maximization
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use performance maximization with classification loss to guide pruning | ResNet-50 · ImageNet | Top-1 accuracy: 75.70% -> 75.96% | Helped |
| use performance maximization with classification loss to guide pruning | ResNet-101 · ImageNet | Top-1 accuracy: +0.46% | Helped |
| use performance maximization with classification loss to guide pruning | MobileNetV2 · ImageNet | Top-1 accuracy: 71.60% -> 72.02% | Helped |
| add performance maximization to differentiable gating (ablation) | ShuffleNetV2+(Small) · ImageNet | Top-1 accuracy: +0.44% | Helped |
| finetune the pruned network rather than train it from scratch (ablation) | NPPM · ImageNet, ShuffleNetV2+(Small) | Top-1 accuracy: 72.32% -> 73.06% | Helped |
| finetune the pruned network rather than train it from scratch (ablation) | NPPM · ImageNet, MobileNetV2 | Top-1 accuracy: 71.14% -> 72.02% | Helped |
| use performance maximization rather than differentiable gating alone (ablation) | ResNet-56 · CIFAR-10 | test accuracy | Helped |
| use gradient modification when combining classification loss and performance maximization (ablation) | NPPM · CIFAR-10, ResNet-56 | test accuracy | Helped |
Closest papers by meaning
- Pruning from Scratch (AAAI 2020 · similarity 0.92)
- Revisiting Random Channel Pruning for Neural Network Compression (CVPR 2022 · similarity 0.91)
- Discrimination-aware Channel Pruning for Deep Neural Networks (NeurIPS 2018 · similarity 0.91)
- Convolutional Neural Network Pruning With Structural Redundancy Reduction (CVPR 2021 · similarity 0.91)
- “Learning-Compression” Algorithms for Neural Net Pruning (CVPR 2018 · similarity 0.91)
- EPSD: Early Pruning with Self-Distillation for Efficient Model Compression (AAAI 2024 · 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.