XAInomaly: Explainable and Interpretable Deep Contractive Autoencoder for O-RAN Traffic Anomaly Detection
Researchers have developed XAInomaly, a new framework utilizing a semi-supervised deep contractive autoencoder for anomaly detection in open radio access networks (O-RAN). This approach aims to learn normal network behavior and identify deviations indicative of anomalies. To overcome the 'black-box' nature of deep learning, the framework incorporates a reactive explainable AI technique called fastshap-C. AI
IMPACT Enhances network management capabilities in O-RAN by providing interpretable anomaly detection.