Artificial intelligence for defense applications

Artificial intelligence (AI) is a set of algorithmic techniques, tools, and technologies that provide machines with the ability to perform tasks that normally require human intelligence – to perceive the world, learn from experience, reason about information, represent knowledge, act, and adapt. Given the multitude of rapid technological advancements in AI, the defense community has emphasized the importance of leveraging these very technologies to be prepared to fight and win the wars of the future. As one of the ways to modernize key capabilities, the defense community has specifically mentioned the need to invest broadly in the military application of AI, including rapid application of commercial breakthroughs, to gain competitive military advantages. To solve some of the most critical problems facing the defense community, the future force requires the ability to converge capabilities from across multiple domains at speeds and scales beyond human cognitive abilities. This special issue is composed of six papers that promote an understanding of AI for defense applications, as well as providing awareness into some of the state-of-theart research and development activities in AI that are applicable to defense applications spanning fraud detection for national security, computer vision for satellite imagery analysis, hidden Markov modeling for the maritime domain, deep learning for radio frequency systems, representation learning for militarily relevant graphs, and robot swarms for military reconnaissance and surveillance. First, the paper by Kerwin and Bastian investigates the national security challenge of predicting fraud, as criminals continually exploit the electronic financial system to defraud consumers and businesses by finding weaknesses in the system, including in audit controls. Their work uses stacked generalizations via meta-learning combined with a resampling methodology particularly useful for the imbalanced fraud data structure to improve fraud detection for national security. Second, the paper by Humphries, Parker, Jonas, Adams, and Clark investigates the problem of quickly and accurately identifying building and road infrastructure via satellite imagery for the execution of tactical military operations in an urban environment. Their work uses an object detection algorithm powered by convolutional neural networks to predict both buildings and road intersections present in an image, as well as use of a contourfinding algorithm for data labeling. Third, the paper by Caelli, Mukerjee, McCabe, and Kirszenblat tackles the problem of integrated sensor and tactical information fusion from a number of sources to enable rapid decision throughput based upon situation awareness for maritime surveillance missions. Their work develops a method using a hidden Markov model to objectively encode, summarize, and analyze airborne maritime surveillance crew activities to gain insights into probabilistic relationships between the attention switching across sensor types and surveyed objects over the entire mission. Fourth, the paper by Clark, Hauser, Headley, and Michaels investigates the radio frequency system problem of automatic modulation classification for situational awareness. Their work examines how useful a synthetically trained system is expected to be when deployed without considering the environment within the synthesis, how training data augmentation can be leveraged for deep learning in the radio frequency domain, and what impact knowledge of degradations to the signal caused by the transmission channel contributes to radio frequency system performance. Fifth, the paper by Lawley, Frey, Mullen and WissnerGross explores the sparse graph representation learning problem for network link prediction and node classification tasks and whole-network reconstruction applicable to militarily relevant graphs such as social and sensor