Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning Robust Global Representations by Penalizing Local Predictive Power
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 |
|---|---|---|---|
| add patch-wise adversarial regularization (PAR) | ResNet · CIFAR-10 perturbed | accuracy: 63.9 -> 66.3 | Helped |
| add patch-wise adversarial regularization (PAR) | AlexNet · ImageNet-Sketch | top-1 accuracy: 0.1204 -> 0.1306 | Helped |
| add patch-wise adversarial regularization (PAR) | AlexNet · PACS | accuracy: 67.03 -> 71.30 | Helped |
| PAR vs domain-ID-using MetaReg | AlexNet · PACS | accuracy: 72.63 -> 72.08 | Hurt |
| MLP local classifier instead of linear (PAR_M) (ablation) | PAR · CIFAR-10 perturbed | accuracy: 66.3 -> 66.5 | No clear effect |
| 3x3 local patches instead of 1x1 (PAR_B) (ablation) | PAR · CIFAR-10 perturbed | accuracy: 66.3 -> 64.2 | Hurt |
| regularize second layer instead of first (PAR_H) (ablation) | PAR · CIFAR-10 perturbed | accuracy: 66.3 -> 62.9 | Hurt |
| regularize second layer instead of first (PAR_H) (ablation) | PAR · PACS | accuracy: 71.30 -> 72.08 | Helped |
| 3x3 local patches instead of 1x1 (PAR_B) (ablation) | PAR · ImageNet-Sketch | top-1 accuracy: 0.1306 -> 0.1273 | Hurt |
Closest papers by meaning
- Robust and Generalizable Visual Representation Learning via Random Convolutions (ICLR 2021 · similarity 0.89)
- SketchNet: Sketch Classification With Web Images (CVPR 2016 · similarity 0.87)
- Adversarially Robust Few-shot Learning via Parameter Co-distillation of Similarity and Class Concept Learners (CVPR 2024 · similarity 0.86)
- Gain from Neighbors: Boosting Model Robustness in the Wild via Adversarial Perturbations Toward Neighboring Classes (CVPR 2025 · similarity 0.85)
- ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object (CVPR 2024 · similarity 0.85)
- LORE: Lagrangian-Optimized Robust Embeddings for Visual Encoders (NeurIPS 2025 · 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.