Carnegie Mellon University’s Cleotilde (Coty) Gonzalez, Christian Lebiere and Lujo Bauer are part of a team that has received a $6.2 million Multidisciplinary University Research Initiative (MURI) grant from the Department of Defense to prevent cyber attacks.
The project, "Realizing Cyber Inception: Towards a Science of Personalized Deception for Cyber Defense," will develop deception tactics based on theories from cognitive science, computational game theory and computer systems engineering. These new tactics are expected to leap ahead of attackers by moving towards active defense, where new cyber environments will make it impossible for attackers to determine what is real and what is deceptive. This new approach to cybersecurity is called "Cyber Inception."
From a multi-pronged standpoint, the researchers will create game theory approaches for modeling defenders and attackers in cyber security environments, build cognitive models of decision-making and develop prototypes for multi-layered environments that can monitor attack attempts and induce deception by taking advantage of human behavior.
We want to be able to trap the adversary in a system that looks entirely real, but is in fact engineered to let us observe the adversary's behavior.Lujo Bauer, Associate Professor, Electrical and Computer Engineering & Institute of Software Research, Carnegie Mellon University
"Informed by basic behavioral science and using our experience in developing computational representations of human decision making processes, we will advance game-theoretic security algorithms to realize cyber inception. This new approach to cybersecurity will exploit the psychology of deception to lure attackers into believing that they have successfully compromised a system, while keeping our systems safe," said Gonzalez, research professor of social and decision sciences in the Dietrich College of Humanities and Social Sciences.
Lebiere, with Gonzalez and colleagues from Arizona State University, will develop computational cognitive models of opponent behavior.
"Our goal is to model the cognitive biases of adversaries to provide a predictive computational basis for game theory models to exploit their attack," said Lebiere, a research psychologist.
Bauer, with colleagues from the University of North Carolina at Chapel Hill and North Carolina State University, will lead the effort to develop computer systems that can convincingly deceive adversaries.
"We want to be able to trap the adversary in a system that looks entirely real, but is in fact engineered to let us observe the adversary's behavior," said Bauer, an associate professor in the College of Engineering and the School of Computer Science.
Milind Tambe, the Helen N. and Emmett Jones Professor of Engineering at the University of Southern California, is the principal investigator on the MURI grant.