10.1007/978-3-032-22211-4_5">
 

MADFACTs: A Meta-learning Augmented Defense Framework for Adversarial Cyber Techniques

Document Type

Conference Proceeding

Publication Date

6-14-2026

Abstract

Adversarial examples are functional data examples that fool machine learning classifiers. Network intrusion detection systems are not typically designed with resilience to adversarial attacks. Recent research demonstrates that generative adversarial attacks allow malicious cyber packets to bypass network intrusion detectors at 69% success rate. This alarming penetration rate opens the door for cyber-kinetic attacks that risk damage to infrastructure and loss of life. A novel framework incorporates meta-learning and adversarial training to enhance the detection of generative adversarial examples. Our results show 100% detection rate against known variants of adversarial attack. Detection of adversarial examples is a critical component of securing cyber networks and in turn protecting infrastructure.

Comments

© The Authors, under exclusive license to Springer Nature Switzerland AG 2026.
The full paper is available from the publisher via subscription or purchase using the DOI link below.

Event: 23rd International Conference, CSC 2025, and 21st International Conference, FCS 2025, Held as Part of the World Congress in Computer Science, Computer Engineering, and Applied Computing, CSCE 2025, Las Vegas, July 21-24, 2025.

Co-authors Chale and Bastian were co-affiliated with the Army Cyber Institute, United States Military Academy at the time of this publication.

Source Publication

Emerging Trends in Scientific Computing and Theoretical Computer Science

This document is currently not available here.

Share

COinS