AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Learning With Multiclass AUC: Theory and Algorithms

Zhiyong Yang, Qianqian Xu, Shilong Bao, Xiaochun Cao, Qingming Huang

TPAMI 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Ranking-based loss optimization · 117 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 exponential-loss multiclass AUC optimization instead of cross-entropydeep model · iNaturalist2017MAUC: 66.84 -> 84.87Improved
use exponential-loss multiclass AUC optimization instead of cross-entropydeep model · CIFAR-100-ImbMAUC: 59.80 -> 62.08Improved
use exponential-loss multiclass AUC optimization instead of cross-entropydeep model · User-ImbMAUC: 55.26 -> 60.94Improved
use squared-loss multiclass AUC optimization instead of DeepAUCdeep model · iNaturalist2017MAUC: 62.48 -> 79.67Improved
use exponential-loss multiclass AUC optimization instead of DeepAUCdeep model · User-ImbMAUC: 59.93 -> 60.94Improved
use hinge loss rather than cross-entropy in the MAUC frameworkdeep model · CIFAR-100-ImbMAUC: 59.80 -> 59.64Hurt
use hinge loss instead of exponential loss in the MAUC frameworkdeep model · iNaturalist2017MAUC: 84.87 -> 79.16Hurt
add BorderlineSMOTE oversampling to CE baselineCE · iNaturalist2017MAUC: 66.84 -> 78.18Improved
add BorderlineSMOTE oversampling to CE baselineCE · CIFAR-100-ImbMAUC: 59.80 -> 59.07Hurt
add TomekLinks undersampling to CE baselineCE · User-ImbMAUC: 55.26 -> 52.99Hurt
add class-balanced reweighting to cross-entropy (CBCE)CE · User-ImbMAUC: 55.26 -> 58.91Improved
add class-balanced reweighting to cross-entropy (CBCE)CE · iNaturalist2017MAUC: 66.84 -> 77.71Improved

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