All news with #anomaly detection tag

Amazon Managed Service for Prometheus Adds Anomaly Detection

🔍 Amazon Managed Service for Prometheus now includes anomaly detection using the Random Cut Forest (RCF) algorithm to continuously analyze time series and surface unexpected metric behavior with minimal user intervention. When you create an anomaly detector in an AMP workspace, it generates four derived time series that represent detected anomalies and their confidence values. Those derived series can be used to build dynamic alerting rules in the AMP Alertmanager and visualized alongside input metrics in self‑managed Grafana or Amazon Managed Grafana. The feature is available in all regions where AMP is generally available and is configurable via the AWS CLI, SDKs, or APIs.

Six Novel Ways to Apply AI in Cybersecurity Defense

🛡️ AI is being applied across security operations in novel ways to predict, simulate, and deter attacks. Experts from BforeAI, NopalCyber, Hughes, XYPRO, AirMDR, and Kontra outline six approaches — predictive scoring, GAN-driven attack simulation, AI analyst assistants, micro-deviation detection, automated triage and response, and proactive generative deception — that aim to reduce alert fatigue, accelerate investigations, and increase attacker costs. Successful deployments depend on accurate ground truth data, continuous model updates, and significant compute and engineering investment.

AWS IoT SiteWise adds automated anomaly model retraining

🔁 AWS announced native anomaly detection enhancements for AWS IoT SiteWise, including automated model retraining, flexible promotion modes, and exposed model metrics. Retraining can be scheduled between 30 days and one year to keep models current with changing equipment conditions. Customers can choose automatic service-managed promotion or manual customer-managed promotion using exposed metrics such as precision, recall, and AUC. Multivariate detection is available in N. Virginia, Ireland, and Sydney.