Locally adaptive one-class classifier fusion with dynamic ℓp-Norm constraints for robust anomaly detection

Sepehr Nourmohammadi, Arda Sarp Yenicesu, Shervin Rahimzadeh Arashloo, Ozgur S. Oguz

Abstract: Highlights•Novel locally adaptive classifier fusion with dynamic constraints for anomaly detection.•New interior-point optimization achieves 19x speedup over traditional fusion approaches.•LiRAnomaly: a comprehensive dataset for robotic manipulation anomaly detection.•Superior results on 12 datasets validated through rigorous statistical analysis.