Research map / Architectures and efficiency / RNNs and normalization
Exploiting Space Folding by Neural Networks
Michal Lewandowski, Raphael Pisoni, Bernhard Heinzl, Bernhard A. Moser
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 |
|---|---|---|---|
| increase total number of neurons in MLP (ablation) | ReLU MLP on MNIST · MNIST | folding sparsity S_N: S_N in (0.4, 0.8) at N=60 -> 0 as N grows | Improved |
| train longer (rising validation accuracy) | ReLU MLP 2x256 · CIFAR100 | global folding Phi_N | Improved |
| add batch norm or dropout (ablation) | small ReLU MLPs · MNIST | global folding Phi_N | No clear effect |
| compare interaction coefficient I for adversarial vs original sample pairs | MNIST MLP · MNIST digits 5 and 7, PGD/FGSM | interaction coefficient I | Mixed |
Closest papers by meaning
- Neural Collapse under Gradient Flow on Shallow ReLU Networks for Orthogonally Separable Data (NeurIPS 2025 · similarity 0.84)
- Adversarial generalization of unfolding (model-based) networks (NeurIPS 2025 · similarity 0.84)
- Generalized Neural Collapse for a Large Number of Classes (ICML 2024 · similarity 0.83)
- Topological obstruction to the training of shallow ReLU neural networks (NeurIPS 2024 · similarity 0.83)
- The Persistence of Neural Collapse Despite Low-Rank Bias (NeurIPS 2025 · similarity 0.83)
- Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse (ICLR 2025 · similarity 0.83)
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.
