Research map / Learning theory and algorithms / Kernels and regression theory
Net-Trim: Convex Pruning of Deep Neural Networks with Performance Guarantee
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 |
|---|---|---|---|
| Apply parallel PCN Net-Trim | MNIST classifier trained with 200 samples · MNIST test | accuracy: 77.5 -> 77.7 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 200 samples · MNIST test, 10% Gaussian noise | accuracy: 46.1 -> 55.0 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 300 samples · MNIST test | accuracy: 82.2 -> 82.6 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 300 samples · MNIST test, 10% Gaussian noise | accuracy: 61.5 -> 70.4 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 500 samples · MNIST test | accuracy: 86.1 -> 86.5 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 500 samples · MNIST test, 10% Gaussian noise | accuracy: 62.1 -> 65.5 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 1000 samples · MNIST test | accuracy: 89.2 -> 89.8 | Helped |
| Apply parallel PCN Net-Trim | MNIST classifier trained with 1000 samples · MNIST test, 10% Gaussian noise | accuracy: 39.2 -> 52.9 | Helped |
| Use cascade rather than parallel Net-Trim | Spiral classifiers · Nested spirals at constant network relative discrepancy | sparsity ratio | Helped |
| Add cascade Net-Trim after dropout and ℓ1-penalty training | Nested-spiral classifier · Nested spirals | relative discrepancy: 0.0183 | No clear effect |
Closest papers by meaning
- Learning to Prune Deep Neural Networks via Layer-wise Optimal Brain Surgeon (NeurIPS 2017 · similarity 0.90)
- Computation and Memory-Efficient Model Compression with Gradient Reweighting (NeurIPS 2025 · similarity 0.90)
- Learning Pruning-Friendly Networks via Frank-Wolfe: One-Shot, Any-Sparsity, And No Retraining (ICLR 2022 · similarity 0.90)
- How Sparse Can We Prune A Deep Network: A Fundamental Limit Perspective (NeurIPS 2024 · similarity 0.89)
- On the Hardness of Training Deep Neural Networks Discretely (AAAI 2025 · similarity 0.88)
- Random Sparse Lifts: Construction, Analysis and Convergence of finite sparse networks (ICLR 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.