A Unified Siamese Learning Framework for Zero-Day Anomaly Detection and Classification in Optical Networks

📰 ArXiv cs.AI

Learn to build a unified Siamese learning framework for zero-day anomaly detection and classification in optical networks with over 99% accuracy

advanced Published 10 Jun 2026

Action Steps

Build a multi-similarity Siamese neural network using Python and TensorFlow to unify anomaly detection and classification
Train the network on a dataset of normal and anomalous traffic patterns in optical networks
Test the network's accuracy and adaptability on unseen anomaly types and lightpaths
Configure the network to detect zero-day anomalies without retraining
Apply the framework to real-world optical networks to improve security and performance

Who Needs to Know This

Network engineers and AI researchers can benefit from this framework to improve optical network security and adaptability

Key Insight

💡 A unified Siamese learning framework can achieve high accuracy and instant adaptability in anomaly detection and classification for optical networks

Key Takeaways

Learn to build a unified Siamese learning framework for zero-day anomaly detection and classification in optical networks with over 99% accuracy

Full Article

Title: A Unified Siamese Learning Framework for Zero-Day Anomaly Detection and Classification in Optical Networks

Abstract:
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.

Read full paper → ← Back to Reads