AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Can Label-Specific Features Help Partial-Label Learning?

Ruo-Jing Dong, Jun-Yi Hang, Tong Wei, Min-Ling Zhang

AAAI 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 3 citations (OpenAlex)

Read the paper

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 changedWhereMeasuredEffect
Use UCL label-specific features with SUREPLL classifier · Controlled UCI datasetsaccuracyImproved
Use UCL label-specific features with SUREPLL classifier · Real-world partial-label datasetsaccuracyMixed
Add an uncertain partition alongside positive and negative partitions (ablation)UCL label-specific feature generation · zoo, glass, yeast, Lost, MSRCv2, MirflickraccuracyImproved
Remove the positive/negative/uncertain partition and cluster all training instances together (ablation)UCL label-specific feature generation · zoo, glass, yeast, Lost, MSRCv2, MirflickraccuracyHurt
Replace three-way partitioning with positive/negative-only partitioning (ablation)UCL label-specific feature generation · zoo, glass, yeast, Lost, MSRCv2, MirflickraccuracyHurt
Combine UCL features with KNN instead of the default SURE classifier (ablation)UCL · Real-world partial-label datasetsaccuracyMixed
Combine UCL features with SVM instead of the default SURE classifier (ablation)UCL · Real-world partial-label datasetsaccuracyMixed
Vary the confidence thresholds and feature-generation parameters across tested ranges (ablation)UCL · Lost, MSRCv2, glass, tmc2007accuracyNo clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy