Research map / Label-efficient and robust learning / Noisy labels
Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrection
Seong Min Kye, Kwanghee Choi, Joonyoung Yi, Buru Chang
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 label correction (ablation) | FasTEN · Clothing1M | top-1 accuracy: 77.07 -> 77.83 | Improved |
| add label correction (ablation) | FasTEN · CIFAR-10 (sym 80% noise) | overall label accuracy: 0.7559 -> 0.7847 | Improved |
| add label correction (ablation) | FasTEN · CIFAR-10 (sym 80% noise) | overall NLL: 1.0057 -> 0.8889 | Improved |
| use efficient single-stage transition matrix estimation instead of baselines | FasTEN · CIFAR-10 (sym 80% noise) | training time (hours): 10.2 -> 1.54 | Improved |
| use FasTEN instead of MSLC at high noise | ResNet-34 · CIFAR-10 (sym 80% noise) | test accuracy: 64.90 -> 79.52 | Improved |
| use FasTEN instead of MSLC at high noise | ResNet-34 · CIFAR-100 (sym 80% noise) | test accuracy: 21.07 -> 37.36 | Improved |
| use FasTEN instead of MSLC | ResNet-34 · CIFAR-100 (asym 20% noise) | test accuracy: 70.86 -> 70.35 | Hurt |
| use FasTEN instead of MLaC | ResNet-34 · CIFAR-10 (asym 40% noise) | test accuracy: 90.26 -> 90.43 | No clear effect |
| use FasTEN instead of MLaC | ResNet-50 · Clothing1M | top-1 accuracy: 75.78 -> 77.83 | Improved |
| use FasTEN instead of DivideMix | ResNet-50 · Clothing1M | top-1 accuracy: 74.76 -> 77.83 | Improved |
Closest papers by meaning
- Extended : Learning with Mixed Closed-set and Open-set Noisy Labels (TPAMI 2022 · similarity 0.92)
- Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learning (NeurIPS 2020 · similarity 0.91)
- Estimation of the Label-Noise Transition Matrix with Performance Guarantees via Selective Classification (NeurIPS 2026 · similarity 0.90)
- Revisiting Interpolation for Noisy Label Correction (AAAI 2025 · similarity 0.89)
- Learning the Latent Causal Structure for Modeling Label Noise (NeurIPS 2024 · similarity 0.89)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.89)
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.
