Research map / Label-efficient and robust learning / Recognition and long-tail
Neuron with Steady Response Leads to Better Generalization
Qiang Fu, Lun Du, Haitao Mao, Xu Chen, Wei Fang, Shi Han, Dongmei Zhang
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 NSR regularization | MLP-3 · MNIST | error rate: 3.09 -> 2.80 | Improved |
| add NSR regularization | MLP-10 · MNIST | error rate: 3.06 -> 1.72 | Improved |
| add NSR regularization | ResNet-18 · CIFAR-10 | error rate: 4.22 -> 3.74 | Improved |
| add NSR regularization | VGG-19 · CIFAR-10 | error rate: 9.19 -> 8.09 | Improved |
| add NSR regularization | ResNet-50 · ImageNet | top-5 error rate: 7.82 -> 6.98 | Improved |
| add NSR regularization | GCN, 2 layers · PubMed | error rate: 12.02 -> 11.92 | No clear effect |
| add NSR regularization | GraphSAGE and GCN, 2-4 layers · PubMed, Amazon-Photo, Amazon-Computers, WikiCS | error rate: GCN 14.01 -> 12.96 | Improved |
| replace NSR by L1 regularization | MLP-4 · MNIST | error rate: 2.29 -> 2.27 | No clear effect |
| replace NSR by Jacobian regularization | MLP-4 · MNIST | error rate: 2.29 -> 2.21 | No clear effect |
| NSR vs L2 regularization | GraphSAGE · Amazon-Computers | error rate: 10.68 -> 10.52 | Improved |
| add NSR on top of Batch Normalization | MLP-4 · MNIST | error rate: 2.22 -> 1.62 | Improved |
| add NSR on top of Dropout | MLP-4 · MNIST | error rate: 2.19 -> 1.64 | Improved |
Closest papers by meaning
- Neuron Activation Coverage: Rethinking Out-of-distribution Detection and Generalization (ICLR 2024 · similarity 0.88)
- Feature Averaging: An Implicit Bias of Gradient Descent Leading to Non-Robustness in Neural Networks (ICLR 2025 · similarity 0.86)
- How Uniform Random Weights Induce Non-uniform Bias: Typical Interpolating Neural Networks Generalize with Narrow Teachers (ICML 2024 · similarity 0.85)
- Convergence Analysis of Newton's Method for Neural Networks in the Overparameterized Limit (NeurIPS 2026 · similarity 0.85)
- Bias of Stochastic Gradient Descent or the Architecture: Disentangling the Effects of Overparameterization of Neural Networks (ICML 2024 · similarity 0.85)
- Data Distributional Properties As Inductive Bias for Systematic Generalization (CVPR 2025 · similarity 0.85)
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.
