Cognitive Information Processing Challenges for Homeland Security.
Paul A. Kogut, Kenneth Kisiel · 2005
This paper describes the goals and emerging results of an ongoing research project in order to solicit input and feedback from the AI for Homeland Security community. The objective is to gain a greater understanding of the performance issues and computational requirements associated with real world AI applications for intelligence analysis and homeland security. These computational requirements will be used to design next generation hardware/software architectures for cognitive information processing. Problem Many AI techniques are notoriously computationally intensive. What are the performance requirements for the application of various AI techniques to real world