Research map / Architectures and efficiency / Pruning and efficient tuning
Global Sparse Momentum SGD for Pruning Very Deep Neural Networks
Read the paperCheck 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 GSM pruning instead of magnitude-based ticket selection | LeNet-300 winning tickets at 60× compression · MNIST | Top1 accuracy: 97.39 -> 98.22 | Helped |
| Use GSM pruning instead of magnitude-based ticket selection | LeNet-5 winning tickets at 125× compression · MNIST | Top1 accuracy: 97.60 -> 99.04 | Helped |
| Use GSM pruning instead of magnitude-based ticket selection | LeNet-5 winning tickets at 300× compression · MNIST | Top1 accuracy: 11.35 -> 98.88 | Helped |
| Add momentum with β=0.98 rather than β=0.90 (ablation) | GSM on DenseNet-40 at 8× compression · CIFAR-10 | ratio of parameters under 1×10^-4: 87.5%; around 150 epochs -> 400 epochs are not enough | Helped |
| Remove activation re-selection after the first iteration (ablation) | GSM on DenseNet-40 at 8× compression · CIFAR-10 | accuracy | Hurt |
| Add activation re-selection at every training iteration (ablation) | GSM on DenseNet-40 at 8× compression · CIFAR-10 | training loss | Helped |
Closest papers by meaning
- Computation and Memory-Efficient Model Compression with Gradient Reweighting (NeurIPS 2025 · similarity 0.90)
- Layerwise Sparse Coding for Pruned Deep Neural Networks with Extreme Compression Ratio (AAAI 2020 · similarity 0.90)
- DominoSearch: Find layer-wise fine-grained N:M sparse schemes from dense neural networks (NeurIPS 2021 · similarity 0.89)
- SPDY: Accurate Pruning with Speedup Guarantees (ICML 2022 · similarity 0.89)
- Pruning Deep Neural Networks from a Sparsity Perspective (ICLR 2023 · similarity 0.89)
- How Sparse Can We Prune A Deep Network: A Fundamental Limit Perspective (NeurIPS 2024 · similarity 0.88)
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.