Research map / Label-efficient and robust learning / Noisy labels
Instance-dependent Label-noise Learning under a Structural Causal Model
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 CausalNL instead of T-Revision | FashionMNIST · IDN-20% | accuracy: 91.58$\pm$0.31 -> 90.84$\pm$0.31 | Hurt |
| Use CausalNL instead of T-Revision | FashionMNIST · IDN-50% | accuracy: 68.99$\pm$1.04 -> 78.19$\pm$1.01 | Helped |
| Use CausalNL instead of MentorNet | SVHN · IDN-20% | accuracy: 94.08$\pm$0.12 -> 94.06$\pm$0.23 | No clear effect |
| Use CausalNL instead of Mixup | SVHN · IDN-50% | accuracy: 68.95$\pm$2.58 -> 85.41$\pm$2.95 | Helped |
| Use CausalNL instead of MentorNet | CIFAR10 · IDN-20% | accuracy: 81.03$\pm$0.12 -> 81.47$\pm$0.32 | Helped |
| Use CausalNL instead of Decoupling | CIFAR10 · IDN-50% | accuracy: 50.43$\pm$2.19 -> 77.39$\pm$1.24 | Helped |
| Use CausalNL instead of MentorNet | CIFAR100 · IDN-20% | accuracy: 38.91$\pm$0.54 -> 41.47$\pm$0.32 | Helped |
| Use CausalNL instead of MentorNet | CIFAR100 · IDN-50% | accuracy: 24.15$\pm$2.31 -> 32.129$\pm$2.23 | Helped |
| Use CausalNL instead of T-Revision | Clothing1M · test | accuracy: 70.97 -> 72.24 | Helped |
Closest papers by meaning
- Learning Causal Transition Matrix for Instance-dependent Label Noise (AAAI 2025 · similarity 0.90)
- Understanding Instance-Level Label Noise: Disparate Impacts and Treatments (ICML 2021 · similarity 0.90)
- Learning the Latent Causal Structure for Modeling Label Noise (NeurIPS 2024 · similarity 0.88)
- On the Role of Label Noise in the Feature Learning Process (ICML 2025 · similarity 0.88)
- Label Noise: Ignorance Is Bliss (NeurIPS 2024 · similarity 0.87)
- Self-cognitive Denoising in the Presence of Multiple Noisy Label Sources (ICML 2024 · 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.