Research map / Label-efficient and robust learning / Noisy labels
Loss factorization, weakly supervised learning and label noise robustness
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 |
|---|---|---|---|
| Use the mean-operator estimator and doubled-sample update instead of vanilla SGD | Logistic-loss SGD · UCI datasets with asymmetric label noise, (p_-, p_+) = (.20,.49) | test error difference vs SGD: Australian $\bm{-.25}$; breast-can. $\bm{-.08}$; diabetes $\bm{-.22}$; heart $\bm{-.15}$; housing $\bm{-.13}$; ionospher | Helped |
| Use the mean-operator estimator and doubled-sample update instead of vanilla SGD | Logistic-loss SGD · UCI datasets with no label noise, (p_-, p_+) = (.00,.00) | test error difference vs SGD: australian $+.01$; diabetes $\bm{-.03}$; sonar $\pm.00$ | No clear effect |
| Estimate the mean operator with the asymmetric-noise correction | Linear-odd-loss empirical risk minimization · Asymmetric label noise | expected empirical l-risk bias | Helped |
Closest papers by meaning
- Losses over Labels: Weakly Supervised Learning via Direct Loss Construction (AAAI 2023 · similarity 0.91)
- Consistent One-vs-All Losses Robust to Misspecification of the Weak Label Transition Model (NeurIPS 2026 · similarity 0.91)
- Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Rates (ICML 2020 · similarity 0.91)
- Label Noise: Ignorance Is Bliss (NeurIPS 2024 · similarity 0.89)
- Tight and Fast Bounds for Multi-Label Learning (ICML 2025 · similarity 0.88)
- Understanding Domain Generalization: A Noise Robustness Perspective (ICLR 2024 · similarity 0.88)
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.