Research map / Architectures and efficiency / CNN design
Improving Generalization of Batch Whitening by Convolutional Unit Optimization
Yooshin Cho, Hanbyel Cho, Youngsoo Kim, Junmo Kim
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 |
|---|---|---|---|
| use proposed Convolutional Unit (whitening before convolution) without linear transform (ablation) | IterNorm · CIFAR-10 (ResNet20) | top-1 test error (%): 9.01 -> 8.29 | Improved |
| use proposed Convolutional Unit (whitening before convolution) without linear transform (ablation) | IterNorm · CIFAR-100 (ResNet56) | top-1 test error (%): 28.35 -> 27.3 | Improved |
| add gamma scaling to the proposed unit (ablation) | IterNorm · CIFAR-100 (ResNet56) | top-1 test error (%): 27.3 -> 27.64 | Hurt |
| use proposed unit with BN keeping gamma and beta (ablation) | BN · CIFAR-10 (ResNet20) | top-1 test error (%): 8.18 -> 8.43 | Hurt |
| replace original unit with proposed unit plus Grouped Shift | IterNorm · CIFAR-10 (ShiftResNet56) | top-1 test error (%): 7.03 -> 6.66 | Improved |
| replace original unit with proposed unit plus Grouped Shift | IterNorm · CIFAR-100 (ShiftResNet56) | top-1 test error (%): 28.38 -> 27.20 | Improved |
| replace original unit with proposed unit plus Grouped Shift | DBN · CIFAR-100 (ShiftResNet56) | top-1 test error (%): 29.87 -> 27.85 | Improved |
| replace original unit with proposed unit at 10x larger learning rate (lr 1.0) | IterNorm · CIFAR-10 (ShiftResNet56) | top-1 test error (%): 8.28 -> 6.04 | Improved |
| replace original unit with proposed unit at 10x larger learning rate (lr 1.0) | IterNorm · CIFAR-100 (ShiftResNet56) | top-1 test error (%): 30.63 -> 25.51 | Improved |
| replace original unit with proposed unit | IterNorm · ImageNet (ShiftNet-A) | top-1 test error (%): 28.25 -> 27.37 | Improved |
| replace original unit with proposed unit | IterNorm · ImageNet (ShiftNet-A-1.5) | top-1 test error (%): 23.87 -> 23.15 | Improved |
| replace BN with IterNorm using proposed unit | ShiftNet-A · CUB-200-2011 transfer | top-1 test error (%): 17.64 -> 14.10 | Improved |
Closest papers by meaning
- An Investigation Into the Stochasticity of Batch Whitening (CVPR 2020 · similarity 0.91)
- Stochastic Whitening Batch Normalization (CVPR 2021 · similarity 0.91)
- Decorrelated Batch Normalization (CVPR 2018 · similarity 0.90)
- Switchable Whitening for Deep Representation Learning (ICCV 2019 · similarity 0.90)
- Iterative Normalization: Beyond Standardization Towards Efficient Whitening (CVPR 2019 · similarity 0.90)
- Group Whitening: Balancing Learning Efficiency and Representational Capacity (CVPR 2021 · similarity 0.90)
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.
