Research map / Label-efficient and robust learning / Noisy labels
GFD-EMVC: Evolutionary Multi-View Classification with Label Noise via Gradient and Feature Dual-Perception
Shuai Li, Xinyan Liang, Yuhua Qian, Li Lv
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 |
|---|---|---|---|
| calibrate detected noisy labels instead of removing them (ablation) | outlier gradient analysis (OGA) · CIFAR-10N Worst | test accuracy (%): 83.72 -> 84.46 | Improved |
| calibrate detected noisy labels instead of removing them (ablation) | OGA · CIFAR-100N Noisy100 | test accuracy (%): 60.99 -> 61.47 | Improved |
| calibrate detected noisy labels instead of removing them (ablation) | OGA · CIFAR-10N Aggregate | test accuracy (%): 91.36 -> 92.08 | Improved |
| train on purified data | EFB-EMVC · MVoxCeleb (40% label noise) | test accuracy (%): 75.06 -> 80.55 | Improved |
| train on purified data | ETMC · YoutubeFace (40% label noise) | test accuracy (%): 52.23 -> 70.96 | Improved |
| train on purified data | Multiplication fusion · NUS (40% label noise) | test accuracy (%): 26.76 -> 62.53 | Improved |
| train on purified data | ETF · Reuters3 (40% label noise) | test accuracy (%): 80.94 -> 81.31 | Improved |
| purify training data before evolutionary search | EMVC on YoutubeFace · fitness ranking of 15 individuals | rank correlation (PC): 0.3857 (original vs purified ordering) | Improved |
| increase clipping ratio kappa from 2.5% to 5% (ablation) | GFD-EMVC · CIFAR-10N Aggregate | test accuracy (%): 91.64 -> 92.36 | Improved |
| increase clipping ratio kappa beyond noise rate (7.5% to 12.5%) (ablation) | GFD-EMVC · CIFAR-10N Aggregate | test accuracy (%): 92.38 -> 92.17 | Hurt |
Closest papers by meaning
- Evolutionary Multi-View Classification via Eliminating Individual Fitness Bias (NeurIPS 2025 · similarity 0.93)
- Noisy Label Calibration for Multi-View Classification (AAAI 2025 · similarity 0.85)
- Learning Discriminative Dynamics with Label Corruption for Noisy Label Detection (CVPR 2024 · similarity 0.84)
- Learning with Open-world Noisy Data via Class-independent Margin in Dual Representation Space (AAAI 2025 · similarity 0.83)
- Label Noise: Ignorance Is Bliss (NeurIPS 2024 · similarity 0.83)
- Double-Layer Hybrid-Label Identification Feature Selection for Multi-View Multi-Label Learning (AAAI 2024 · similarity 0.83)
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.
