<ul><li>Network Intrusion Detection Systems (NIDS) are crucial for protecting digital infrastructures from cyber threats.</li><li>ODXU, a Neurosymbolic AI framework, enhances NIDS robustness, interpretability, and generalization by integrating deep embedded clustering, symbolic reasoning with XGBoost, and uncertainty quantification (UQ).</li><li>ODXU outperforms traditional neural models in various evaluation metrics based on experimental results on the CIC-IDS-2017 dataset.</li><li>A transfer learning strategy is developed to reuse a pre-trained ODXU model on a different dataset, showing improved performance with minimal training samples.</li></ul>

Neurosymbolic Artificial Intelligence for Robust Network Intrusion Detection: From Scratch to Transfer Learning

Discover more