Research map / Architectures and efficiency / CNN design
Beyond Dropout: Feature Map Distortion to Regularize Deep Neural Networks
Yehui Tang, Yunhe Wang, Yixing Xu, Boxin Shi, Chao Xu, Chunjing Xu, Chang Xu
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 |
|---|---|---|---|
| Replace RDdrop with feature map distortion on fully connected layers | CNN · CIFAR-10 | accuracy: 83.11 -> 85.24 | Improved |
| Replace RDdrop with feature map distortion on fully connected layers | CNN · CIFAR-100 | accuracy: 54.65 -> 56.23 | Improved |
| Replace DropBlock with block feature map distortion | ResNet-56 · CIFAR-10 | accuracy: 94.18 -> 94.50 | Improved |
| Replace DropBlock with block feature map distortion | ResNet-56 · CIFAR-100 | accuracy: 73.08 -> 73.71 | Improved |
| Replace Dropout with feature map distortion | ResNet-50 · ImageNet validation | top-1 accuracy: 76.80 to 77.71 | Improved |
| Replace DropBlock with block feature map distortion | ResNet-50 · ImageNet validation | top-1 accuracy: 78.13 $\pm$ 0.05 -> 78.76 $\pm$ 0.05 | Improved |
| Use block feature map distortion rather than elementwise feature map distortion | ResNet-50 · ImageNet validation | top-1 accuracy: 77.71 $\pm$ 0.05 -> 78.76 $\pm$ 0.05 | Improved |
| Increase distortion probability or distortion intensity | Feature map distortion · CIFAR-100 | test accuracy | Mixed |
| Increase distortion probability or distortion intensity | Feature map distortion · CIFAR-100 | accuracy gap between training and testing | Improved |
Closest papers by meaning
- Adaptive Dropout with Rademacher Complexity Regularization (ICLR 2018 · similarity 0.88)
- Demystifying Dropout (ICML 2019 · similarity 0.87)
- Dropout with Expectation-linear Regularization (ICLR 2017 · similarity 0.87)
- Guided Dropout (AAAI 2019 · similarity 0.87)
- Shakeout: A New Regularized Deep Neural Network Training Scheme (AAAI 2016 · similarity 0.87)
- Shakeout: A New Approach to Regularized Deep Neural Network Training (TPAMI 2017 · 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.
