A Unified Siamese Learning Framework for Zero-Day Anomaly Detection and Classification in Optical Networks
arXiv:2606.10827v1 Announce Type: cross Abstract: A multi-similarity Siamese neural network unifies zero-day anomaly detection and one-shot classification in optical networks, achieving over 99% accuracy and instant adaptability across lightpaths and unseen anomaly types without any retraining.
The increasing complexity and criticality of optical networks, coupled with the rising sophistication of cyber threats and anomalies, necessitates advanced, real-time security solutions like this one.
This development significantly enhances the security and resilience of critical digital infrastructure by enabling instant, adaptive detection of novel threats without human intervention or retraining.
Optical networks can now autonomously identify and classify previously unseen anomalies with high accuracy, reducing response times and operational overhead compared to traditional methods.
- · Telecommunications companies
- · Network security providers
- · Critical infrastructure operators
- · AI/ML in cybersecurity sector
- · Traditional signature-based intrusion detection systems
- · Manual network security operations
Reduced downtime and improved security posture for optical network operators.
Accelerated adoption of AI-driven autonomous security solutions across various critical infrastructure sectors.
Potential for an arms race between AI-powered anomaly detection and increasingly sophisticated, AI-generated network attacks.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI