AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Boosting Out-of-distribution Detection with Typical Features

Yao Zhu, YueFeng Chen, Chuanlong Xie, Xiaodan Li, Rong Zhang, Hui Xue', Xiang Tian, bolun zheng, Yaowu Chen

NeurIPS 2022 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 16 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
Apply BATS feature rectification with Energy scoringResNet-50 · ImageNet-1k OOD averageFPR95: 58.16 -> 27.11Improved
Apply BATS feature rectification with Energy scoringResNet-50 · ImageNet-1k OOD averageAUROC: 85.82 -> 94.28Improved
Apply BATS feature rectification with Energy scoringResNet-18 · CIFAR-10 OOD averageFPR95: 32.99 -> 29.96Improved
Apply BATS feature rectification with Energy scoringResNet-18 · CIFAR-100 OOD averageFPR95: 58.12 -> 52.73Improved
Apply BATS feature rectificationEnergy score on ImageNet-1k, ResNet-50 · ImageNet-1k OOD averageFPR95: 58.16 -> 27.11Improved
Apply BATS feature rectificationGradNorm on ImageNet-1k, ResNet-50 · ImageNet-1k OOD averageFPR95: 40.08 -> 21.53Improved
Add feature rectification (ablation)ResNet-50 Energy detector · ImageNet-1k OOD averageFPR95: 58.16 -> 27.11Improved
Rectify Block1 only instead of using no rectification (ablation)ResNet-50 Energy detector · ImageNet-1k OOD averageFPR95: 58.16 -> 59.08Hurt
Rectify Block2 only instead of using no rectification (ablation)ResNet-50 Energy detector · ImageNet-1k OOD averageFPR95: 58.16 -> 57.62Improved
Rectify Block3 only instead of using no rectification (ablation)ResNet-50 Energy detector · ImageNet-1k OOD averageFPR95: 58.16 -> 53.84Improved
Rectify all blocks rather than Block4 only (ablation)ResNet-50 Energy detector · ImageNet-1k OOD averageFPR95: 27.11 -> 26.12Improved
Use a very small truncation threshold (ablation)BATS · ImageNet-1k and CIFAR-100 OOD detectionFPR95Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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