AI Paper Atlas

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

JMLR 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Anomaly detection with contaminated data · 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
solve primal SSLM with global solver Gurobi instead of local solver Snoptprimal SSLM · Banana Scenario Coptimal value: 12024.7000 -> 1747.4100Improved
solve primal SSLM with global solver Gurobi instead of local solver Snoptprimal SSLM · Banana Scenario Boptimal value: 1586.8200 -> 121.9522Improved
use global solver Gurobi instead of local Snoptprimal SSLM · Banana Scenario Aoptimal value: 1.3246 -> -225.8368Improved
start Snopt from (0,1,0) instead of (0,0,0)primal SSLM with Snopt · Banana Scenario Coptimal value: 12024.7000 -> 1747.4100Improved
solve the dual SSLM instead of the nonconvex primaldual SSLM · Banana Scenario Coptimal value: 1747.41 vs 1733.1000Hurt
replace SSLM by convex CSSLMprimal CSSLM · Banana Scenario Boptimal value: 121.7294 -> 121.9522Improved
CSSLM versus SVM in anomaly detectionCSSLM · BTSC (AD)G-means (%): 48.78 -> 62.04Improved
CSSLM versus SVM in anomaly detectionCSSLM · BCW (AD)G-means (%): 41.90 -> 81.91Improved
CSSLM versus SVM in anomaly detectionCSSLM · KDD-Cup'99 (AD)G-means (%): 81.19 -> 87.81Improved
CSSLM versus SVM in binary classificationCSSLM · BCW (BC)G-means (%): 91.62 -> 83.90Hurt
CSSLM versus SVM in binary classificationCSSLM · BTSC (BC)AUPRC: 0.8954 -> 0.8681Hurt
CSSLM versus SVM in anomaly detectionCSSLM · MC (AD)AUPRC: 0.9910 -> 0.9881Hurt

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