Research map / Architectures and efficiency / RNNs and normalization
Self-Normalizing 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 SELU-based self-normalizing networks instead of competing FNN methods | FNNs · UCI 121 classification datasets | average rank difference: -0.240* -> -0.756 | Helped |
| Use SNN instead of SVM | machine-learning methods · UCI datasets with more than 1,000 data points | average rank difference: -6.4 -> -6.7 | Helped |
| Use SNN instead of the strongest FNN baseline at 8 layers | FNNs · Tox21 challenge test set, 8 layers | AUC: 81.9 ± 1.2 -> 84.5 ± 0.2 | Helped |
| Increase network depth from 2 to 8 layers | SNN · Tox21 challenge test set | AUC: 83.7 ± 0.3 -> 84.5 ± 0.2 | Mixed |
| Use SNN instead of MSRAinit | FNNs · HTRU2 10-fold cross-validation | AUC: 0.9791 ± 0.010 -> 0.9803 ± 0.010 | No clear effect |
Closest papers by meaning
- Beyond BatchNorm: Towards a Unified Understanding of Normalization in Deep Learning (NeurIPS 2021 · similarity 0.88)
- Differentiable Learning-to-Normalize via Switchable Normalization (ICLR 2019 · similarity 0.87)
- Batch normalization is sufficient for universal function approximation in CNNs (ICLR 2024 · similarity 0.87)
- On the Nonlinearity of Layer Normalization (ICML 2024 · similarity 0.86)
- Transformers without Normalization (CVPR 2025 · similarity 0.85)
- Scalable Neural Network Kernels (ICLR 2024 · 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.