In response to the pranks, which have been happening nearly every day in recent months, the Coast Guard is planning to adopt voice recognition software to identify the phony callers. While these challenges make it hard to eliminate fake callers, voice recognition may be able to catch the pranksters as the Coast Guard believes most of the calls originate from a small number of callers. It's not entirely clear if the Coast Guard's adoption of voice recognition technology will provide any relief from the pranksters; many software systems record an extended conversation -- up to 40 seconds of talking -- to ID a voice accurately, and callers can potentially thwart the system just by disguising how they talk. The voice recognition software likely wouldn't give the Coast Guard what it needs to make such an arrest or obtain a warrant but would be a solution that would mitigate the damage of the calls without leading to jail time for the callers.
Hoax callers are a special kind of jerk. At best they can cause their targets emotional distress. At worst they can cause or incite property destruction and even divert scarce resources from real emergencies where they are needed to save lives. Technology can sometimes be applied to catch the criminal. Doing so casts technology specialists in the detective role by letting them trace calls to find their origins.
An, Bo (University of Southern California) | Shieh, Eric (University of Southern California) | Tambe, Milind (University of Southern California) | Yang, Rong (University of Southern California) | Baldwin, Craig (United States Coast Guard) | DiRenzo, Joseph (United States Coast Guard) | Maule, Ben (United States Coast Guard) | Meyer, Garrett (United States Coast Guard)
While three deployed applications of game theory for security have recently been reported, we as a community of agents and AI researchers remain in the early stages of these deployments; there is a continuing need to understand the core principles for innovative security applications of game theory. PROTECT is premised on an attacker-defender Stackelberg game model and offers five key innovations. First, this system is a departure from the assumption of perfect adversary rationality noted in previous work, relying instead on a quantal response (QR) model of the adversary's behavior --- to the best of our knowledge, this is the first real-world deployment of the QR model. Fourth, our experimental results illustrate that PROTECT's QR model more robustly handles real-world uncertainties than a perfect rationality model.