Research map / Label-efficient and robust learning / Recognition and long-tail
Fast Multi-Instance Multi-Label Learning
Sheng-Jun Huang, Wei Gao, Zhi‐Hua Zhou
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 shared low-dimensional space W0 across labels (ablation) | MIMLfast · Scene | average precision: .730 -> .770 | Improved |
| add shared low-dimensional space W0 across labels (ablation) | MIMLfast · Reuters | average precision: .963 -> .972 | Improved |
| add shared low-dimensional space W0 across labels (ablation) | MIMLfast · Scene | hamming loss: .211 -> .188 | Improved |
| add dummy label for choosing relevant labels instead of top-r (ablation) | MIMLfast · Scene | hamming loss: .196 -> .188 | Improved |
| add dummy label for choosing relevant labels instead of top-r (ablation) | MIMLfast · Reuters | hamming loss: .035 -> .028 | Improved |
| add dummy label for choosing relevant labels instead of top-r (ablation) | MIMLfast · Scene | average precision: .767 -> .770 | No clear effect |
| add dummy label for choosing relevant labels instead of top-r (ablation) | MIMLfast · Reuters | average precision: .971 -> .972 | No clear effect |
| use K=10 sub-concepts per label instead of K=1 (ablation) | MIMLfast · Scene | average precision: .754 -> .777 | Improved |
| use K=10 sub-concepts per label instead of K=1 (ablation) | MIMLfast · Scene | ranking loss: .209 -> .184 | Improved |
| use K=10 sub-concepts per label instead of K=1 (ablation) | MIMLfast · Reuters | average precision: .972 -> .976 | Improved |
| increase K from 10 to 15 (ablation) | MIMLfast · Scene | average precision: .777 -> .769 | No clear effect |
Closest papers by meaning
- Scalable Optimization of Multivariate Performance Measures in Multi-instance Multi-label Learning (AAAI 2017 · similarity 0.93)
- T-MDML: Triplet-based Multiple Distance Metric Learning for Multi-Instance Multi-Label Classification with Label Correlation (AAAI 2025 · similarity 0.89)
- Fast Multi-Instance Partial-Label Learning (AAAI 2025 · similarity 0.89)
- Multi-Instance Multi-Label Classification from Crowdsourced Labels (AAAI 2025 · similarity 0.88)
- Multi-Instance Partial-Label Learning with Margin Adjustment (NeurIPS 2024 · similarity 0.86)
- Enhancing Multi-Label Classification via Dynamic Label-Order Learning (AAAI 2024 · similarity 0.85)
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.
