Research map / Optimization / Implicit bias and generalization
Weight Conditioning for Smooth Optimization of Neural Networks
Hemanth Saratchandran, Thomas X Wang, Simon Lucey
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 equilibrated weight conditioning to batch norm | Inception · CIFAR100 | Top-1 accuracy (%): 55.9 -> 58.7 | Improved |
| add equilibrated weight conditioning to batch norm | Inception · CIFAR10 | Top-1 accuracy (%): 83.1 -> 83.3 | No clear effect |
| add equilibrated weight conditioning to batch norm | DenseNet · CIFAR10 | Top-1 accuracy (%): 70.5 -> 79.96 | Improved |
| add equilibrated weight conditioning to batch norm | DenseNet · CIFAR100 | Top-1 accuracy (%): 64.6 -> 72.4 | Improved |
| use equilibrated weight conditioning instead of weight standardization | DenseNet + BN · CIFAR10 | Top-1 accuracy (%): 75.9 -> 79.96 | Improved |
| add equilibrated weight conditioning to layernorm | ViT-B · ImageNet1k | Top-1 accuracy (%): 79.9 -> 80.2 | No clear effect |
| apply equilibrated weight conditioning to NeRF MLP | NeRF · LLFF Fern | PSNR (dB): 27.51 -> 28.53 | Improved |
| apply equilibrated weight conditioning to NeRF MLP | NeRF · LLFF Fortress | PSNR (dB): 33.16 -> 33.14 | No clear effect |
| apply equilibrated weight conditioning to NeRF MLP | NeRF · LLFF (8 scenes) | PSNR (dB): 29.6 -> 30. 20 | Improved |
Closest papers by meaning
- Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks (NeurIPS 2016 · similarity 0.90)
- Centered Weight Normalization in Accelerating Training of Deep Neural Networks (ICCV 2017 · similarity 0.89)
- Implicit Regularization and Convergence for Weight Normalization (NeurIPS 2020 · similarity 0.89)
- Understanding the Disharmony between Weight Normalization Family and Weight Decay (AAAI 2020 · similarity 0.89)
- Optimization Theory for ReLU Neural Networks Trained with Normalization Layers (ICML 2020 · similarity 0.88)
- Deep Weight Factorization: Sparse Learning Through the Lens of Artificial Symmetries (ICLR 2025 · similarity 0.86)
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.
