Research map / Label-efficient and robust learning / OOD and anomaly detection
Outlier Detection for Robust Multi-Dimensional Scaling
Leonid Blouvshtein, Daniel Cohen‐Or
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add triangle-inequality outlier filtering before MDS | SMACOF · Motion data | Silhouette: 0.282 -> 0.321 | Improved |
| add triangle-inequality outlier filtering before MDS | SMACOF · MPEG7 | Silhouette: 0.38 -> 0.48 | Improved |
| add triangle-inequality outlier filtering before MDS | SMACOF · 3D mesh | Calinski-Harabasz: 50.7 -> 59.0 | Improved |
| add triangle-inequality outlier filtering before MDS | SMACOF · MPEG7 | AMI: 0.78 -> 0.85 | Improved |
| add triangle-inequality outlier filtering before MDS | SMACOF · Motion data | AMI: 0.64 -> 0.66 | Improved |
| use TMDS instead of FG12 on 1070db shapes | MDS embedding · 1070db 100 shapes | Silhouette: 0.19 -> 0.27 | Improved |
| add TMDS filtering to SMACOF on 1070db shapes | SMACOF · 1070db 100 shapes | Silhouette: 0.14 -> 0.27 | Improved |
| raise outlier rate above 22% with TMDS filtering (ablation) | TMDS · synthetic uniform hypercube | embedding score | Hurt |
| subsample 45 triangles per edge instead of all triangles (ablation) | TMDS · synthetic, 15 outliers | outliers detected: 45 triangles per edge | No clear effect |
| use TMDS instead of FG12 for speed | outlier-robust MDS · synthetic, N=900 | CPU time (s): 491.4 -> 59.9 | Improved |
| increase log-normal sigma of distance distortion (ablation) | TMDS vs SMACOF · log-normal distortion | embedding score | Improved |
| use Sammon weighting for shortened outliers | SMACOF · synthetic outliers | embedding quality | Hurt |
Closest papers by meaning
- Reliable Neighborhood-Aware Multi-View Outlier Detection (ICML 2026 · similarity 0.88)
- Provable Self-Representation Based Outlier Detection in a Union of Subspaces (CVPR 2017 · similarity 0.87)
- Latent Discriminant Subspace Representations for Multi-View Outlier Detection (AAAI 2018 · similarity 0.87)
- RODEO: Robust Outlier Detection via Exposing Adaptive Out-of-Distribution Samples (ICML 2024 · similarity 0.85)
- Towards Generalizable 3D Anomaly Detection via Relational Inconsistency Modeling (NeurIPS 2026 · similarity 0.84)
- Coreset for Robust Geometric Median: Eliminating Size Dependency on Outliers (NeurIPS 2025 · similarity 0.84)
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.
