🛡️ One-Class SVM Anomaly Boundary: RBF Gamma and Nu
A 3D decision surface of a one-class SVM with an RBF kernel: tune gamma and nu, see the boundary, support vectors and flagged anomalies, with six checks (KKT, nu-property, strong duality, kernel-density limit).
⚙ Under the hood
A 3D decision surface of a one-class SVM with an RBF kernel: tune gamma and nu, see the boundary, support vectors and flagged anomalies, with six checks (KKT, nu-property, strong duality, kernel-density limit).
anomaly-detectionone-class-svmrbf-kernelmachine-learningnovelty-detection
3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install