Defense Advanced Research Projects AgencyTagged Content List

Cyber

Relating to digital systems and information

Showing 14 results for Cyber + Algorithms RSS
March 6-7, 2019,
Hilton Alexandria Mark Center
DARPA has long been a leader in the field of artificial intelligence, establishing the foundations of the field and leading creation of expert systems, and then supporting the expansion of machine learning. The agency’s most recent investments — undertaken as part of DARPA’s $2 billion AI Next campaign — are supporting a shift in AI systems from tools alone to trusted, collaborative partners in problem solving. To increase awareness of DARPA’s expansive AI R&D efforts, the agency is hosting an Artificial Intelligence Colloquium (AIC) in March 2019. The event will bring together the Department of Defense research community and stakeholders to learn more about DARPA’s current and emerging AI programs, and discover how the technologies in development could apply to diverse missions.
The growing sophistication and ubiquity of machine learning (ML) components in advanced systems dramatically expands capabilities, but also increases the potential for new vulnerabilities. Current research on adversarial AI focuses on approaches where imperceptible perturbations to ML inputs could deceive an ML classifier, altering its response.
LADS will develop a new protection paradigm that separates security-monitoring functionality from the protected system, focusing on low-resource, embedded and Internet of Things (IoT) devices. The program will explore technologies to associate the running state of a device with its involuntary analog emissions across different physical modalities including, but not limited to, electromagnetic emissions, acoustic emanations, power fluctuations and thermal output variations.
Researchers have demonstrated effective attacks on machine learning (ML) algorithms. These attacks can cause high-confidence misclassifications of input data, even if the attacker lacks detailed knowledge of the ML classifier algorithm and/or training data. Developing effective defenses against such attacks is essential if ML is to be used for defense, security, or health and safety applications.
Serial Interactions in Imperfect Information Games Applied to Complex Military Decision Making (SI3-CMD) builds on recent developments in artificial intelligence and game theory to enable more effective decisions in adversarial domains. SI3-CMD will explore several military decision making applications at strategic, tactical, and operational levels and develop AI/game theory techniques appropriate for their problem characteristics.