Research paper on applying pre-trained time-series models to cross-machine anomaly detection, combining transfer learning with distributed system monitoring.
Research
Cross-Machine Anomaly Detection Leveraging Pre-trained Time-series Model
Transfer learning from pre-trained time-series models enables anomaly detection across heterogeneous machines without per-system retraining, reducing distributed infrastructure monitoring overhead.
Wednesday, April 8, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.LG (Machine Learning)BY sys://pipeline
Tags
research