Research map / Probabilistic and scientific modeling / Variational and generative sampling
Multi-view Anomaly Detection via Robust Probabilistic Latent Variable Models
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| infer multiple latent vectors per instance versus PCCA (one latent vector) (ablation) | Proposed · 11 LIBSVM data sets | AUC: best on eight of the 11 | Helped |
| increase anomaly rate | all methods · 11 LIBSVM data sets | AUC | Hurt |
| consensus clustering (CC) with 20 clusters | CC · 11 LIBSVM data sets | AUC | Hurt |
| single-view method OCSVM on multi-view anomalies | OCSVM · 11 LIBSVM data sets | AUC | Hurt |
| latent dimensionality K=1 or 2 instead of larger (ablation) | Proposed · anomaly rate 0.4 | AUC | Hurt |
| latent dimensionality above four (ablation) | all methods · anomaly rate 0.4 | AUC | No clear effect |
| Student-t marginalization of noise precision (Gamma prior) (ablation) | Proposed · Synthetic single-view anomalies | AUC: 0.860 -> 0.117 | Helped |
| PCCA versus proposed on single-view anomalies | PCCA · Synthetic single-view anomalies | AUC: 0.174 vs 0.117 | No clear effect |
| Gibbs sampling vs PCCA computation | Proposed · wine data | computation time (s): 2 -> 35 | Hurt |
| HOAD hyperparameter grid {0.1, 1, 10, 100}, best per data set (ablation) | HOAD · 11 LIBSVM data sets | AUC | Mixed |
Closest papers by meaning
- Multi-View Anomaly Detection: Neighborhood in Locality Matters (AAAI 2019 · similarity 0.88)
- Unveiling Multi-View Anomaly Detection: Intra-view Decoupling and Inter-view Fusion (AAAI 2025 · similarity 0.85)
- Robust Multi-view Clustering against Imperfect Information (NeurIPS 2026 · similarity 0.83)
- Beyond Normal References: Discriminative Few-Shot Anomaly Detection (NeurIPS 2026 · similarity 0.83)
- Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity (CVPR 2025 · similarity 0.83)
- One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection (NeurIPS 2024 · similarity 0.83)
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.