Principal Investigators
- Michael Wellman (U Michigan)
- Satinder Singh (U Michigan)
- Demosthenis Teneketzis (U Michigan)
PIs at partner universities:
- Sushil Jajodia (George Mason U) [Project Director]
- George Cybenko (Dartmouth U)
- Peng Liu (Pennsylvania State U)
This project is funded as part of the MURI program by the Army Research Office under grant W911NF-13-1-0421.
Project Summary
Today’s cyber defenses are largely static. They are governed by slow deliberative processes involving testing, security patch deployment, and human-in-the-loop monitoring. As a result, adversaries can systematically probe target networks, pre-plan their attacks, and ultimately persist for long times inside compromised networks and hosts. A new class of technologies, called Adaptive Cyber Defense (ACD), is being developed that presents adversaries with optimally changing attack surfaces and system configurations, forcing adversaries to continually re-assess and re-plan their cyber operations. Although these approaches (e.g., moving target defense, dynamic diversity, and bio-inspired defense) are promising, they assume stationary and stochastic, but non-adversarial, environments. To realize the full potential, we need to build the scientific foundations so that system resiliency and robustness in adversarial settings can be rigorously defined, quantified, measured, and extrapolated in a rigorous and reliable manner.
Related Publications
- Empirical game-theoretic methods for adaptive cyber-defense
- Probably almost-stable strategy profiles in simulation-based games
- Iterated Deep Reinforcement Learning in Games: History-Aware Training for Improved Stability
- Deception in Finitely Repeated Security Games
- A Learning and Masking Approach to Secure Learning
- Stackelberg Security Games: Looking Beyond a Decade of Success
- Multi-stage attack graph security games: Heuristic strategies, with empirical game-theoretic analysis
- A Stackelberg game model for botnet data exfilitration
- Adversarial and Uncertain Reasoning for Adaptive Cyber Defense: Building the Scientific Foundation
- SoK: Security and Privacy in Machine Learning
- A Moving Target Defense Approach to Mitigate DDoS Attacks against Proxy-Based Architectures
- Moving Target Defense against DDoS Attacks: An Empirical Game-Theoretic Analysis
- Gradient Methods for Stackelberg Security Games
- Empirical Game-Theoretic Analysis for Moving Target Defense
- Empirical Game-Theoretic Analysis of an Adaptive Cyber-Defense Scenario (Preliminary Report)