Research map / Label-efficient and robust learning / Domain adaptation and generalization
Regularizing CNN Transfer Learning With Randomised Regression
Check 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 |
|---|---|---|---|
| Add pseudo-task regularization to fine-tuning | VGG-16 transfer learning · Flower102 | accuracy gain: 2.38% (0.32) | Helped |
| Add pseudo-task regularization to fine-tuning | VGG-16 transfer learning · CUB200 | accuracy gain: 3.05% (0.39) | Helped |
| Add pseudo-task regularization to fine-tuning | VGG-16 transfer learning · MIT67 | accuracy gain: 1.42% (0.58) | Helped |
| Replace SML1 with L2 pseudo-regression loss (ablation) | PtR with VGG-16 · Flower102, CUB200, MIT67, Stanford40, WebFace500 | accuracy gain: 2.38% (0.32) vs 2.61% (0.42); 3.05% (0.39) vs 2.84% (0.37); 1.42% (0.58) vs 1.39% (0.40); 2.50% (0.09) vs 2.21% (0.16); | No clear effect |
| Disable weight decay (ablation) | PtR with ResNet-50 · CUB200 | accuracy: 81.9 to 82.0 | Helped |
| Disable weight decay (ablation) | PtR · Flower102 | accuracy gain | No clear effect |
| Use PtR instead of fine-tuning alone | ResNet-50 transfer learning · CUB200 | accuracy gain: 1.6 | Helped |
| Use PtR instead of fine-tuning alone | ResNet-101 transfer learning · Caltech256-30 | accuracy gain: 0.5 | Helped |
| Use FNP instead of fine-tuning alone | ResNet-101 transfer learning · Caltech256-30 | accuracy gain: -0.2 | Hurt |
Closest papers by meaning
- Combining Explicit and Implicit Regularization for Efficient Learning in Deep Networks (NeurIPS 2022 · similarity 0.88)
- Network as Regularization for Training Deep Neural Networks: Framework, Model and Performance (AAAI 2020 · similarity 0.88)
- Regularizing Deep Networks by Modeling and Predicting Label Structure (CVPR 2018 · similarity 0.87)
- MetaPerturb: Transferable Regularizer for Heterogeneous Tasks and Architectures (NeurIPS 2020 · similarity 0.87)
- Demystifying Batch Normalization in ReLU Networks: Equivalent Convex Optimization Models and Implicit Regularization (ICLR 2022 · similarity 0.87)
- Explicit Inductive Bias for Transfer Learning with Convolutional Networks (ICML 2018 · 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.