Aris Kanellopoulos, “Secure Autonomous Cyber-Physical Systems in Adversarial Environments”
Aris Kanellopoulos
Research Engineer
Georgia Institute of Technology
Title: Secure Autonomous Cyber-Physical Systems in Adversarial Environments
Abstract: Autonomous cyber-physical systems have been increasingly introduced to various facets of society. Those, often large-scale, platforms reside equally in the physical and the cyber domains. Their exposition to the full complexity of human-centric environments coupled with their own intrinsic complexity renders them extremely vulnerable to attacks from adversarial agents, both of machine and human nature. As such, it is of the utmost importance to develop security mechanism that are able to respond to the adaptability and unpredictability of the attackers in order to predict, detect and mitigate attacks on such critical systems. In our work, we propose methods of rendering cyber-physical systems safe and secure, by exploiting their structure via novel defense mechanisms, inspired by principles of computer security, reinforcement learning and game theory. Furthermore, we investigate realistic adversarial models based on the ideas from behavioral economics, brought into the rigorous mathematical framework of control systems.
Kavli Room, Clark Hall 3rd floor
or on Zoom:
https://wse.zoom.us/j/6886507095
One tap mobile: US: +13017158592,,6886507095# or +16465588656,,6886507095#
Meeting URL: https://wse.zoom.us/j/6886507095?from=addon
Meeting ID: 688 650 7095
Join by Telephone
For higher quality, dial a number based on your current location.
Dial:
US: +1 301 715 8592 or +1 646 558 8656 or +1 312 626 6799 or +1 253 215 8782 or +1 346 248 7799 or +1 669 900 6833
Meeting ID: 688 650 7095