Research map / Label-efficient and robust learning / Recognition and long-tail
Progressive Identification of True Labels for Partial-Label Learning
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 progressive, seamless label-weight updates rather than iterative EM-style updates (ablation) | PRODEN · UCI and real-world partial-label datasets | test accuracy | Helped |
| Keep label weights uniformly distributed instead of updating them (ablation) | PRODEN · MNIST, Fashion-MNIST, Kuzushiji-MNIST, and CIFAR-10 | test accuracy | Hurt |
| Concentrate all label weight on the current winner at every iteration (ablation) | PRODEN · MNIST, Fashion-MNIST, Kuzushiji-MNIST, and CIFAR-10 | test accuracy | Hurt |
| Replace seamless weight updates with periodic EM-style updates (ablation) | PRODEN · Synthetic Control (binomial, q=0.7) | test accuracy: 95.00±4.75% -> 48.33±7.96% | Hurt |
| Replace seamless weight updates with periodic EM-style updates (ablation) | PRODEN · Lost | test accuracy: 76.57±1.47% -> 68.09±3.78% | Hurt |
| Replace seamless weight updates with periodic EM-style updates (ablation) | PRODEN · Dermatology (binomial, q=0.7) | test accuracy: 95.34±3.15% -> 76.22±4.48% | Hurt |
| Use progressive label identification instead of the competing methods | Partial-label classifiers · Real-world partial-label datasets | test accuracy | Mixed |
| Use PRODEN rather than GA or D²CNN | Deep partial-label classification · UCI and real-world partial-label datasets | test accuracy | Helped |
| Increase ambiguity degree under pair flipping | PRODEN and baseline methods · Kuzushiji-MNIST | test accuracy | Mixed |
Closest papers by meaning
- Realistic Evaluation of Deep Partial-Label Learning Algorithms (ICLR 2025 · similarity 0.91)
- What Makes Partial-Label Learning Algorithms Effective? (NeurIPS 2024 · similarity 0.90)
- Conformal Prediction for Partial Label Learning (AAAI 2025 · similarity 0.90)
- CroSel: Cross Selection of Confident Pseudo Labels for Partial-Label Learning (CVPR 2024 · similarity 0.89)
- Partial Label Learning with a Partner (AAAI 2024 · similarity 0.89)
- Reduction-based Pseudo-label Generation for Instance-dependent Partial Label Learning (NeurIPS 2025 · 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.