AI for Safety and Security

Presented by Arunesh Sinha

Thursday, February 9, 2016
2:00 p.m.
ICSI Lecture Hall

Abstract:

In any security setting, defenders are beset with the problem of limited resources. Optimal allocation of limited defense resources is an important issue for many defense agencies. Recognizing the fact that a defender-adversary interaction is inherently a game, I will present my past work on game theoretic models for optimal defense resource allocation in an adversarial environment. I will talk about two projects: a project in collaboration with TSA that introduced a game model for screening of passengers and provided novel optimization methods to compute the equilibrium and another project that provided a machine learning based crime prediction tool for the University of Southern California police. I will also briefly mention about my ongoing work on adversarial learning.

Speaker Bio:

Arunesh Sinha is an Assistant Research Scientist in the Computer Science and Engineering Department at the University of Michigan.