ROME, N.Y. — The assistant secretary of the Army for acquisition, logistics, and technology - ASA (ALT) selected ANDRO Computation Solutions, LLC for a phase I Small Business Innovation Research (SBIR) award.
The Rome–based company will use the money to develop DeepGenerator, a technology for rapidly generating realistic and diverse tactical-data sets for various customer-specific applications to enhance their deployed machine-learning (ML) models. This is the next in a series of multiple contract awards from the U.S. Department of Defense, totaling more than $20 million for ANDRO.
The award extends the ANDRO AI Innovation Lab’s portfolio of generative artificial intelligence (AI) solutions by addressing the key challenge of data scarcity in efficiently training ML systems. Principal Investigator Jithin Jagannath, Sean Furman, Anu Jagannath, Chris Maracchion, and Haolun Li lead the lab team.
It’s often expensive or even impossible to collect mission- or application-relevant data in many practical deployment situations and under diverse operating conditions. DeepGenerator will bridge the gap with a focus on computer-vision applications, counter-uncrewed air system, and security and threat detection, where more representative data, synthetically generated, are needed to improve the performance of core AI solutions. ANDRO will use funding under the next phases to incrementally expand and mature DeepGenerator’s capabilities and support transitions to the warfighter domain.