AI Paper Atlas

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

Conditional Bernoulli Mixtures for Multi-label Classification

Cheng Li, Bingyu Wang, Virgil Pavlu, Javed Aslam

ICML 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 32 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 CBM mixture instead of independent binary relevance (LR)logistic regression · MEDIAMILLsubset accuracy: 9.6 -> 13.5Improved
Use CBM instead of BinRel (LR)logistic regression · SCENEsubset accuracy: 51.5 -> 69.7Improved
Use CBM instead of PowSet (LR)logistic regression · RCV1subset accuracy: 50.2 -> 49.9Hurt
Use CBM instead of PowSet (LR)logistic regression · MEDIAMILLsubset accuracy: 9.0 -> 13.5Improved
Swap LR base learner for gradient boosted trees in CBMCBM · MEDIAMILLsubset accuracy: 13.5 -> 14.1Improved
Swap LR base learner for gradient boosted trees in CBMCBM · SCENEsubset accuracy: 69.7 -> 70.5Improved
Swap LR base learner for GB in CBMCBM · RCV1subset accuracy: 49.9 -> 43.0Hurt
Swap LR for GB in BinRelBinRel · RCV1subset accuracy: 40.4 -> 30.1Hurt
Swap LR for GB in BinRelBinRel · SCENEsubset accuracy: 51.5 -> 59.3Improved
Swap LR for GB in PowSetPowSet · TMC2007subset accuracy: 28.2 -> 23.1Hurt
Increase number of mixture components K from 1 to 15 (ablation)CBM · TMCsubset accuracyImproved
Increase K beyond about 30 (ablation)CBM · TMCsubset accuracyNo 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