AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Hyperparameter Sensitivity in Deep Outlier Detection: Analysis and a Scalable Hyper-Ensemble Solution

Xueying Ding, Lingxiao Zhao, Leman Akoglu

NeurIPS 2022 · Architectures and efficiency › RNNs and normalization · approach: Dataset distillation and data-efficient training · 6 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
Average scores across models with different hyperparameter configurations rather than selecting a single configurationDeep OD models · MNIST-4, CleanAUROC: 92.39 -> 95.94Improved
Average scores across all hyperparameter configurations rather than selecting a single modelDeep OD models · MNIST-4, PollutedAUROC: 85.73 -> 90.46Improved
Use RobOD instead of RandNetOutlier detection · MNIST-4AUROC: 85.3$\pm$3.1 -> 88.0$\pm$0.0Improved
Use RobOD instead of the strongest deep baselineOutlier detection · MNIST-5AUROC: 79.3$\pm$3.8 -> 81.4$\pm$0.1Improved
Use RobOD instead of the strongest deep baselineOutlier detection · MNIST-8AUROC: 85.4$\pm$2.2 -> 87.8$\pm$0.0Improved
Use RobOD instead of the strongest baselineOutlier detection · CardioAUROC: 94.1$\pm$0.9 -> 93.5$\pm$0.1Hurt
Use RobOD instead of the strongest baselineOutlier detection · ThyroidAUROC: 97.9$\pm$0.4 -> 86.1$\pm$0.6Hurt
Use RobOD instead of the strongest baselineOutlier detection · LymphoAUROC: 99.5$\pm$0.3 -> 98.7$\pm$0.1Hurt
Train each member on 10% of the data rather than using full-data RobOD (ablation)RobOD · MNIST-8AUROC: 87.8$\pm$0.0 -> 88.2$\pm$0.1Improved
Train each member on 10% of the data rather than using full-data RobOD (ablation)RobOD · MNIST-5AUROC: 81.4$\pm$0.1 -> 78.5$\pm$0.1Hurt
Train each member on 10% of the data rather than using full-data RobOD (ablation)RobOD · ThyroidAUROC: 86.1$\pm$0.6 -> 88.4$\pm$0.5Improved
Train ensemble members independently rather than using shared-parameter RobOD (ablation)RobOD · MNIST-4AUROC: 88.0$\pm$0.0 -> 84.5$\pm$0.0Hurt

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