AI Paper Atlas

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

OpenAUC: Towards AUC-Oriented Open-Set Recognition

Zitai Wang, Qianqian Xu, Zhiyong Yang, Yuan He, Xiaochun Cao, Qingming Huang

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Learning from unlabeled and positive-unlabeled data · 9 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
add the close-set correctness switch (OpenAUC coupling) to the AUC risk instead of plain Acc+AUC (ablation)OpenAUC learning method · CIFAR10OpenAUC: 87.6 -> 89.2Improved
add the close-set correctness switch (OpenAUC coupling) to the AUC risk instead of plain Acc+AUC (ablation)OpenAUC learning method · TinyImageNetOpenAUC: 74.0 -> 75.9Improved
add the close-set correctness switch (OpenAUC coupling) to the AUC risk instead of plain Acc+AUC (ablation)OpenAUC learning method · MNISTOpenAUC: 99.3 -> 99.4Improved
add the close-set correctness switch (OpenAUC coupling) to the AUC risk instead of plain Acc+AUC (ablation)OpenAUC learning method · CIFAR+50OpenAUC: 92.0 -> 93.6Improved
replace CE+ training with OpenAUC risk optimizationVGG32 CE+ · TinyImageNetOpenAUC: 74.3 -> 75.9Improved
replace CE+ training with OpenAUC risk optimizationVGG32 CE+ · CIFAR10OpenAUC: 88.1 -> 89.2Improved
replace Softmax with OpenAUC risk optimizationVGG32 · SVHNOpenAUC: 92.8 -> 95.0Improved
replace CE+ with OpenAUC risk optimizationResNet50 on CUB · CUB HardAUC: 76.3 -> 78.1Improved
replace CE+ with OpenAUC risk optimizationResNet50 on CUB · CUB close-setclose-set accuracy: 86.2 -> 86.2No clear effect
increase lambda weight of AUC risk beyond 0.1-0.2 (ablation)OpenAUC learning method · benchmarks (Fig sensitivity)OpenAUCHurt

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