CCICADA Seminar Series in Homeland Security

The Challenges and Opportunities in Game Theory for Security

Date/Time: Friday, April 19, 2024, 11:45 am to 1:30 pm
Location:  4th Floor Lounge – 401, Computing Research & Education Building (CoRE), Busch Campus, Rutgers, the State University

***Light lunch served at 12 noon, talk at 12:15 pm. Please RSVP to Nicole Clark-Johnson <nicolec@dimacs.rutgers.edu> if you will be attending lunch.***

FEATURED SPEAKER: Arunesh Sinha, Assistant Professor in the Department of Management Science & Information Systems, Rutgers Business School at Rutgers University

Abstract:
Game theory is a very relevant tool to model defender-adversary interaction in physical security problems. However, similar to any other models, game theoretic models are not perfect but are useful. In this talk, I will present an evolution in modeling security games using the Stackelberg game model, from models that use standard game theoretic assumptions (rationality, known utilities) to ones that learn adversary behavior from data and then optimize against the learned behavior. The three line of work that I will present span from standard game theory model for prevention of terrorist attacks (domain with very little data) to data-driven defense strategy design in poaching prevention (domain with data) to reinforcement learning based design of threat screening policies for a zero-sum game model (using a simulator). A common challenge in all these scenarios is scalability, which has resulted in novel contributions in optimization in general. Finally, I will discuss some fundamental challenges in extending the game theoretic ideas to cyber-security.

Bio: Arunesh Sinha is an Assistant Professor in the Department of Management Science & Information Systems, Rutgers Business School at Rutgers University. Prior to this, he was an Assistant Professor in the School of Computing and Information Systems at Singapore Management University. Prior to the position in Singapore, he was a Research Scientist at the University of Michigan and a postdoctoral scholar at the University of Southern California. He received his Ph.D. from Carnegie Mellon University and his undergraduate degree from IIT Kharagpur, India. Dr. Sinha has conducted research at the intersection of security, machine learning and game theory. His interests lie in the theoretical aspects of multi-agent interaction, machine learning, security and privacy, along with an emphasis on the real-world applicability of the theoretical models. His papers have won awards at the AAMAS and ACSAC conference and he was invited for the IJCAI Early Career Spotlight talk at the IJCAI 2023 conference.

***Please RSVP to Nicole Clark-Johnson <nicolec@dimacs.rutgers.edu> if you will be attending lunch.

Leave a comment

Your email address will not be published.


*


This site uses Akismet to reduce spam. Learn how your comment data is processed.