Research map / Label-efficient and robust learning / Noisy labels
CA2C: A Prior-Knowledge-Free Approach for Robust Label Noise Learning via Asymmetric Co-learning and Co-training
Mengmeng Sheng, Zeren Sun, Tianfei Zhou, Xiangbo Shu, Jinshan Pan, Yazhou Yao
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 asymmetric co-learning with paradigm deconstruction (ACPD) (ablation) | CA2C · CIFAR100N Sym-20% | accuracy: 35.14 -> 65.22 | Improved |
| add asymmetric co-training with cross-guidance label generation (ACLG) (ablation) | CA2C · CIFAR100N Sym-20% | accuracy: 65.22 -> 67.37 | Improved |
| add confidence-based re-weighting (CBRW) (ablation) | CA2C · CIFAR100N Sym-20% | accuracy: 67.37 -> 68.64 | Improved |
| add asymmetric co-learning with paradigm deconstruction (ACPD) (ablation) | CA2C · CIFAR100N Sym-80% | accuracy: 4.41 -> 28.53 | Improved |
| add asymmetric co-training with cross-guidance label generation (ACLG) (ablation) | CA2C · CIFAR100N Sym-80% | accuracy: 28.53 -> 35.90 | Improved |
| add confidence-based re-weighting (CBRW) (ablation) | CA2C · CIFAR100N Sym-80% | accuracy: 35.90 -> 40.97 | Improved |
| add asymmetric co-learning with paradigm deconstruction (ACPD) (ablation) | CA2C · CIFAR100N Asym-40% | accuracy: 27.29 -> 62.02 | Improved |
| add asymmetric co-training with cross-guidance label generation (ACLG) (ablation) | CA2C · CIFAR100N Asym-40% | accuracy: 62.02 -> 65.06 | Improved |
| add confidence-based re-weighting (CBRW) (ablation) | CA2C · CIFAR100N Asym-40% | accuracy: 65.06 -> 65.59 | Improved |
| add ACPD (ablation) | CA2C · CIFAR80N Sym-80% | accuracy: 4.20 -> 28.01 | Improved |
| add ACLG (ablation) | CA2C · CIFAR80N Asym-40% | accuracy: 60.39 -> 64.63 | Improved |
| set candidate label set size K to 2 (ablation) | CA2C · CIFAR100N Sym-20% | accuracy: highest performance at K = 2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- UNICON — UniCon: Combating Label Noise Through Uniform Selection and Contrastive Learning (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.91)
- Safeguarded Dynamic Label Regression for Noisy Supervision (AAAI 2019 · similarity 0.90)
- BadLabel: A Robust Perspective on Evaluating and Enhancing Label-Noise Learning (TPAMI 2024 · similarity 0.90)
- Mitigating Label Noise through Data Ambiguation (AAAI 2024 · similarity 0.88)
- NoiseGPT: Label Noise Detection and Rectification through Probability Curvature (NeurIPS 2024 · similarity 0.88)
- Label Noise: Ignorance Is Bliss (NeurIPS 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.
