Research map / Label-efficient and robust learning / OOD and anomaly detection
Towards Convexity in Anomaly Detection: A New Formulation of SSLM with Unique Optimal Solutions
Hongying Liu, Hao Wang, Haoran Chu, Yibo Wu
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 |
|---|---|---|---|
| solve primal SSLM with global solver Gurobi instead of local solver Snopt | primal SSLM · Banana Scenario C | optimal value: 12024.7000 -> 1747.4100 | Improved |
| solve primal SSLM with global solver Gurobi instead of local solver Snopt | primal SSLM · Banana Scenario B | optimal value: 1586.8200 -> 121.9522 | Improved |
| use global solver Gurobi instead of local Snopt | primal SSLM · Banana Scenario A | optimal value: 1.3246 -> -225.8368 | Improved |
| start Snopt from (0,1,0) instead of (0,0,0) | primal SSLM with Snopt · Banana Scenario C | optimal value: 12024.7000 -> 1747.4100 | Improved |
| solve the dual SSLM instead of the nonconvex primal | dual SSLM · Banana Scenario C | optimal value: 1747.41 vs 1733.1000 | Hurt |
| replace SSLM by convex CSSLM | primal CSSLM · Banana Scenario B | optimal value: 121.7294 -> 121.9522 | Improved |
| CSSLM versus SVM in anomaly detection | CSSLM · BTSC (AD) | G-means (%): 48.78 -> 62.04 | Improved |
| CSSLM versus SVM in anomaly detection | CSSLM · BCW (AD) | G-means (%): 41.90 -> 81.91 | Improved |
| CSSLM versus SVM in anomaly detection | CSSLM · KDD-Cup'99 (AD) | G-means (%): 81.19 -> 87.81 | Improved |
| CSSLM versus SVM in binary classification | CSSLM · BCW (BC) | G-means (%): 91.62 -> 83.90 | Hurt |
| CSSLM versus SVM in binary classification | CSSLM · BTSC (BC) | AUPRC: 0.8954 -> 0.8681 | Hurt |
| CSSLM versus SVM in anomaly detection | CSSLM · MC (AD) | AUPRC: 0.9910 -> 0.9881 | Hurt |
Closest papers by meaning
- We Need to Rethink Benchmarking in Anomaly Detection (NeurIPS 2026 · similarity 0.85)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.84)
- TailedCore: Few-Shot Sampling for Unsupervised Long-Tail Noisy Anomaly Detection (CVPR 2025 · similarity 0.84)
- Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection (CVPR 2025 · similarity 0.84)
- Towards a Unified Framework of Clustering-based Anomaly Detection (ICML 2025 · similarity 0.83)
- Long-Tailed Anomaly Detection with Learnable Class Names (CVPR 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.
