US Air Force teams with UK on machine learning demo

WASHINGTON — The Air Force Research Laboratory successfully demonstrated new joint machine learning algorithms with the United Kingdom, showcasing the countries’ ability to collaborate on deploying artificial intelligence in support of war fighters.

The cooperation between the Air Force Research Laboratory and the U.K.’s Defense Science and Technology Laboratory (Dstl) is part of a four-year Autonomy and Artificial Intelligence Collaboration (AAIC) Partnership Agreement signed in December 2020. While AFRL is the lead agency for the partnership, the Office of the Under Secretary for Research and Engineering, the Navy and the Army are also participating. This was the first event in a series of planned demonstrations.

The demonstration took place simultaneously at AFRL’s Information Directorate in New York and Dstl’s Salisbury location in the U.K. on Oct. 18. During the event, the two countries simulated a combat scenario with U.K. and U.S. forces adjacent to one another. Using a common platform, operators were able to share data and machine learning algorithms to support wide-area situational awareness, giving brigades better information for decision making.

“The October 18 event demonstrated how the U.K. and U.S. can integrate AI technology to create the first end-to-end Machine Learning (ML) research, development, and deployment ecosystem enabling rapid data sharing, algorithm development, evaluation, and deployment. AI will play a critical role in accelerating decision making to meet the pace & scale of the future battlespace,” said AFRL project lead Lee Seversky.

The U.K. was able to showcase its model cards, which allow commanders to rapidly understand and select the best machine learning algorithms to apply to a mission. The U.S. was able to use the government-owned StreamlinedML, an open platform where users can build machine learning workflows, evaluate models and then deploy them. AFRL said the two countries used 15 machine learning algorithms, 12 data sets and five automated machine learning workflows during the event.

“This collaboration with AFRL & the U.S. services is crucial to drive the very latest AI technology into military operations and innovative research in both nations. The demonstration is just the first step toward our ambition of deploying novel AI that can learn in the field into an experimental trial environment, something that hasn’t been done before and is only possible due to this collaboration,” said Todd Robinson, who heads up the U.K.’s involvement in the partnership.

Nathan Strout covers space, unmanned and intelligence systems for C4ISRNET.