The Marine Corps seeks to employ advanced artificial intelligence (AI) technologies for its CAC2S program to reduce information overload, improve situational awareness (SA) and collaboration, and aid in Commander decision-making. Leverege will use its unrivaled subject matter expertise in CAC2S and its deep experience with cutting-edge machine learning techniques to research and test the optimal algorithms for the AI-based digital assistant. Our research and development will identify the areas that need to be addressed, propose and study alternative solutions, and provide simulated but relevant input to test our hypotheses against known CAC2S use cases. Our proposed technical solution is a modified version of Googles SyntaxNet and TensorFlow; an open-source AI framework that implements a feedforward neural network with a greedy transition-based parser. The open-source nature of those frameworks allows us to ultimately build an embedded AI system that can be installed on a classified CAC2S network without a connection to the Internet.
Benefit: Leverege intends to address the following quantitative commercialization metrics and achieve increased maturity in the following areas: Multi-level neural network approaches to speech tokenization, deep machine learning, and contextual understanding, Conversational bot technology that remembers 0x9D context between individual queries, Autonomous big data 0x9D architectures and implementation strategies that can be embedded into disconnected systems and meet stringent performance and security requirements, and Improved data presentation and user experience methodologies for digital assistants. Commercial applications of the research and development include customer service, self-driving cars, image recognition, decision making, recommendations, and scheduling.
Keywords: Artificial Intelligence, Artificial Intelligence, Machine Learning, CAC2S, Predictive Analytics, Decision Aid, Big Data, Command and Control, Digital Assistant
