AI Paper Atlas

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

Confidence-Rated Discriminative Partial Label Learning

Cai-Zhi Tang, Min-Ling Zhang

AAAI 2017 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 80 citations (OpenAlex)

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 changedWhereMeasuredEffect
use confidence-rated boosting (CORD) instead of uniform-weight convex approach CLPLCLPL · Lostinductive accuracy: 0.742 -> 0.806Improved
CORD vs CLPLCLPL · Yahoo! Newsinductive accuracy: 0.462 -> 0.624Improved
CORD vs PL-KNNPL-KNN · Soccer Playerinductive accuracy: 0.497 -> 0.457Hurt
CORD vs LSB-CMMLSB-CMM · Soccer Playerinductive accuracy: 0.525 -> 0.457Hurt
CORD vs LSB-CMMLSB-CMM · Yahoo! Newsinductive accuracy: 0.648 -> 0.624Hurt
CORD vs PL-SVMPL-SVM · MSRCv2inductive accuracy: 0.482 -> 0.474No clear effect
CORD vs PL-KNN across controlled UCI configsPL-KNN · UCI controlled setswin/tie/loss: 53/34/81Mixed
CORD vs CLPL across controlled UCI configsCLPL · UCI controlled setswin/tie/loss: 135/33/0Improved
CORD vs PL-SVM and LSB-CMM across controlled UCI configsPL-SVM, LSB-CMM · UCI controlled setswin/tie/loss: 114/26/28; 88/68/12Improved
predict labels from confidence matrix instead of classifier output (ablation)CORD · Real-world setstransductive accuracy: 0.764 -> 0.763 (Soccer Player)No clear effect
CORD transductive vs best baselinesCLPL, PL-KNN, PL-SVM, LSB-CMM · Losttransductive accuracy: 0.894 -> 0.925Improved

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