AI Paper Atlas

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

Improvements on Uncertainty Quantification for Node Classification via Distance Based Regularization

Russell Alan Hart, Linlin Yu, Yifei Lou, Feng Chen

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Uncertainty and misclassification detection · 0 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
re-tune entropy regularization weight using validation cross-entropy (ablation)GPN · Citeseer OOD detectionAUROC epistemic with propagation: 70.74 -> 73.75Improved
tune activation function (ReLU, Sigmoid, GELU) in addition (ablation)GPN-CE · Citeseer OOD detectionAUROC epistemic with propagation: 73.75 -> 77.97Improved
add distance-based regularization (ablation)GPN-CE-ACT · Citeseer OOD detectionAUROC epistemic with propagation: 77.97 -> 78.98Improved
re-tune entropy regularization weight using validation cross-entropy (ablation)GPN · PubMed OOD detectionAUROC epistemic with propagation: 73.91 -> 78.32Improved
tune activation function in addition (ablation)GPN-CE · PubMed OOD detectionAUROC epistemic with propagation: 78.32 -> 78.32No clear effect
add distance-based regularization (ablation)GPN-CE-ACT · PubMed OOD detectionAUROC epistemic with propagation: 78.32 -> 81.76Improved
add distance-based regularization (ablation)GPN-CE-ACT · PubMed OOD detectionAUPR epistemic with propagation: 74.11 -> 78.16Improved
re-tune entropy weight on validation CE (ablation)GPN · PubMedID accuracy: 94.08 -> 93.84Hurt
use full method instead of GPNGPN · CoraML misclassification detectionAUPR aleatoric: 49.51 -> 89.95Improved
use full method instead of GPNGPN · CoraML misclassification detectionAUROC aleatoric: 81.19 -> 75.8Hurt
use full method instead of GPNGPN · Citeseer misclassification detectionAUROC aleatoric: 75.89 -> 69.15Hurt
remove propagation layer (w/o network)Ours · CoraML OOD detectionAUROC epistemic: 87.20 -> 76.12Hurt

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