Technical staff members in the Intelligence and Decision Technologies Group review imagery of a location of interest in the maps projected in the PED Lab.

Artificial Intelligence Software Architectures and Algorithms

The Artificial Intelligence (AI) Software Architectures and Algorithms Group is striving to lead the nation in applying AI and machine learning technologies to meet critical national security needs. We are focusing on applications that address counterterrorism, counter human-trafficking, humanitarian assistance and disaster response, protection of the homeland, and global security. In collaboration with industry and academia, we are working at the frontiers of AI to develop prototypes that can help solve national security problems affecting the land, sea, air, and space domains. Our technology portfolio ranges from applications that use AI to augment human capability to those that enable autonomous systems. Our close-knit research teams have access to large amounts of data and apply modern coding practices to identify and execute highly impactful projects. Our staff exploit a deep understanding of relevant operational datasets and of the workflows of intelligence analysts, military commanders, and law enforcement to develop useful tools for these users. Our technical expertise spans deep learning, computer vision, natural language processing, graph theory, big data architectures, operations research, and system autonomy.

Featured Projects

This image illustrates a retweet network for the #MacronLeaks narrative during the 2017 French presidential election. Each circle represents a Twitter account, and arrows represent “retweets.”
New software takes advantage of artificial intelligence techniques to automate the detection and assessment of aggressive influence operations.
Simulation of a flight test
For experiments with unmanned aerial vehicles, researchers use simulated geographies that avoid the risks of real-world flight tests.
Graphic of TbD-net subtasks
A new neural network model reveals some of the mystery behind the decision-making capability of artificial intelligence systems.
A goal of the program is to develop a system that can identify spatial relationships between objects in a scene, such as counting how many planes are parked at the terminal on the left.
An artificial intelligence application that automatically identifies objects in aerial imagery could reduce the time analysts spend in manually combing through images.

Advancing Our Research

Featured Publications

Automated posterior interval evaluation for inference in probabilistic programming

Oct 22
Intl. Conf. on Probabilistic Programming, PROBPROG, 22 October 2020. triangle counting performance [e-print]

Sep 22
2020 IEEE High Performance Computing Conf., HPEC, 22-24 September 2020.

Leveraging linear algebra to count and enumerate simple subgraphs

Sep 22
2020 IEEE High Performance Extreme Computing Conf., HPEC, 22-24 September 2020.