Metrics
8 Record Views
Conference proceeding
Uncertainty-Quantified Neurosymbolic AI for Open Set Recognition in Network Intrusion Detection
MILCOM IEEE Military Communications Conference, pp.13-18
MILCOM 2024 (Washington, DC, USA, 10/28/2024–11/01/2024)
10/28/2024
Web of Science ID: WOS:001419571100053
Abstract
Network Intrusion Detection Systems (NIDS) are crucial for safeguarding networks by detecting and classifying malicious traffic in real-time. This paper presents a novel neurosymbolic artificial intelligence (NSAI) approach for enhancing NIDS which we call ODXU, which combines neural networks with symbolic reasoning to improve classification accuracy, particularly for challenging classes. Additionally, we introduce uncertainty quantification techniques-Confidence Scoring, Shannon Entropy, and post-hoc Uncertainty Metamodeling-to enhance the reliability of the NIDS. Our experimental results demonstrate that our NSAI model, coupled with post-hoc Uncertainty Meta-modeling, outperforms traditional methods, providing superior detection accuracy and robust uncertainty estimates.
Related links
Details
- Title
- Uncertainty-Quantified Neurosymbolic AI for Open Set Recognition in Network Intrusion Detection
- Publication Details
- MILCOM IEEE Military Communications Conference, pp.13-18
- Resource Type
- Conference proceeding
- Conference
- MILCOM 2024 (Washington, DC, USA, 10/28/2024–11/01/2024)
- Publisher
- IEEE; 10.1109/MILCOM61039.2024
- Number of pages
- 6
- Grant note
- U.S. Military Academy (10.13039/100009923) Defense Advanced Research Projects Agency (10.13039/100000185) Arm (10.13039/100016311)
- Copyright
- © 2024, IEEE
- Identifiers
- WOS:001419571100053; 99381512425206600
- Academic Unit
- Intelligent Systems and Robotics; Hal Marcus College of Science and Engineering
- Language
- English