AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Provable Self-Representation Based Outlier Detection in a Union of Subspaces

Chong You, Daniel P. Robinson, Rene Vidal

CVPR 2017 · Graph learning › Graph algorithms and clustering · approach: Graph embedding and spectral sparsification · 99 citations (OpenAlex)

Check 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 R-graph instead of Outlier PursuitOutlier detection · Extended Yale B, one subject, 35% outliersAUC: 0.972 -> 0.986Improved
Use R-graph instead of Outlier PursuitOutlier detection · Extended Yale B, three subjects, 15% outliersAUC: 0.968 -> 0.985Improved
Use R-graph instead of LRROutlier detection · Caltech-256, one category, 50% outliersAUC: 0.907 -> 0.948Improved
Use R-graph instead of L1-thresholdingOutlier detection · Caltech-256, three categories, 50% outliersAUC: 0.810 -> 0.929Improved
Use R-graph instead of L1-thresholdingOutlier detection · Caltech-256, five categories, 50% outliersAUC: 0.774 -> 0.913Improved
Use R-graph instead of L1-thresholdingOutlier detection · Coil-100, one category, 50% outliersAUC: 0.991 -> 0.997Improved
Use R-graph instead of L1-thresholdingOutlier detection · Coil-100, four categories, 25% outliersF1: 0.941 -> 0.970Improved
Use R-graph instead of L1-thresholdingOutlier detection · Coil-100, seven categories, 15% outliersF1: 0.897 -> 0.955Improved
Use self-representation connection patterns rather than representation-vector magnitudeOutlier detection on Extended Yale B face images · Extended Yale Boutlier detection performanceImproved
Increase the number of inlier categoriesOutRank and CoP · Caltech-256AUCHurt
Vary alpha over the tested range [1, 50]R-graph · Extended Yale B, three inlier groups and 15% outliersperformanceNo clear effect
Vary the percentage of outliers from 1% to 15%R-graph · Extended Yale B, three inlier groupsperformanceNo clear effect

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