Communications of the ACM
Machine Learning Algorithm Boosts Defense Against Zero-Day Attacks
Credit: Getty Images
A Penn State-led team of researchers has created an automatically adaptive cyber defense against attacks on vulnerable computer networks and cyber-infrastructure using a machine learning algorithm based on a technique known as reinforcement learning.
The team developed this approach to address what it saw as limitations in a method to detect and respond to cyber-attacks called moving target defense (MTD), according to Minghui Zhu, associate professor of electrical engineering and computer science at Penn State.
The researchers describe their work in "Adaptive Cyber Defense Against Multi-Stage Attacks Using Learning-Based POMDP," published in the ACM Transactions on Privacy and Security.
The team's adaptive manual target-defense techniques "can dynamically and proactively reconfigure deployed defenses that can increase uncertainty and complexity for attackers during vulnerability windows," says Zhu.
From Insurance Journal
View Full Article
No entries found