Research map / Label-efficient and robust learning / Recognition and long-tail
Does the Data Processing Inequality Reflect Practice? On the Utility of Low-Level Tasks
Roy Turgeman, Tom Tirer
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 pre-classification denoising (DnCNN) before classifier | ResNet18 · Noisy CIFAR-10 (sigma 0.25, 0.4) | probability of error | Improved |
| lower noise level (higher SNR) for denoising pre-processing (ablation) | ResNet18 + DnCNN · Noisy CIFAR-10 | maximal relative error reduction | Improved |
| add self-supervised 256-d encoding before classifier | ResNet50 + MLP · Noisy Mini-ImageNet (sigma 50/255, 100/255) | probability of error | Improved |
| lower noise level (higher SNR) for encoding pre-processing (ablation) | ResNet50 + MLP · Noisy Mini-ImageNet | maximal relative error reduction | Improved |
| use encoding instead of denoising as pre-classification step (ablation) | ResNet18 · Noisy CIFAR-10 | relative error reduction | Improved |
Closest papers by meaning
- Optimal Binary Classification Beyond Accuracy (NeurIPS 2022 · similarity 0.87)
- Demystifying the Optimal Performance of Multi-Class Classification (NeurIPS 2023 · similarity 0.85)
- Leveraging Data Symmetries to Select an Optimal Subset of Training Data under Label Noise (NeurIPS 2026 · similarity 0.84)
- Restoring balance: principled under/oversampling of data for optimal classification (ICML 2024 · similarity 0.84)
- DRoP: Distributionally Robust Data Pruning (ICLR 2025 · similarity 0.84)
- Classes Are Not Equal: An Empirical Study on Image Recognition Fairness (CVPR 2024 · similarity 0.84)
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.
