NewsBrief November 12, 2021

Federal data scientists try to find essential truths in a big and messy sea of data

(Federal News Network) The COVID-19 pandemic produced a bumper crop of almost everything it touched, including research. One authoritative database reports that 4% of all scientific research published last year had to do with COVID. And, you guessed it, that produced data. Lots and lots of data. Now the Pacific Northwest National Laboratory is trying to find meaning in a murky sea of data. With more on the project, data scientist Neeraj Kumar joined Federal Drive with Tom Temin. Read More

Missile Defense: Recent Acquisition Policy Changes Balance Risk and Flexibility, but Actions Needed to Refine Requirements Process

(GAO@100) In 2002, the Department of Defense (DOD) provided the Missile Defense Agency (MDA) with flexibilities to diverge from traditional requirements-setting and acquisition processes and instead implement a unique approach to managing its acquisitions. After completing studies in 2019, DOD revised those flexibilities in 2020 by making significant changes to MDA’s requirements-setting and acquisition processes (see figure). Most notably, the Under Secretary of Defense for Acquisition and Sustainment, rather than the MDA Director, now determines whether major MDA programs may progress through the development phases. Read More

CMMC 2.0 to pare down cybersecurity requirements for contractors

(fedscoop) The Department of Defense’s cybersecurity compliance program for contractors will be pared down in scope and expectations, according to an acquisition regulation document. The Cybersecurity Maturity Model Certification (CMMC) will no longer require every contractor to get a third-party certification if they do not touch controlled unclassified data, a change that could reduce the cost of compliance for thousands of contractors. Under the new CMMC model, which is known as CMMC 2.0, the number of security tiers is being shrunk from five to three. Novel CMMC maturity practices will also be eliminated from the standard. Read More

Navy UAS demo displays potential for future cargo resupply

(NAVAIR) NAVAL AIR WARFARE CENTER AIRCRAFT DIVISION, ST. INIGOES, Md. – The Navy and Marine Corps may soon have a way to resupply warfighters on the front line using unmanned systems capable of delivering necessary supplies from a distance in a variety of scenarios. The Department is considering several cargo resupply systems and recently demonstrated these capabilities to senior leadership during an event Oct. 27 at NAWCAD’s Webster Outlying Field. The Navy and Marine Corps Small Tactical Unmanned Aircraft Systems program office (PMA-263) and Naval Air Warfare Center Aircraft Division coordinated this event where operators from Air Test and Evaluation Squadron Two Four (UX-24) performed multiple resupply missions. Read More

Balancing data security and open access remains a challenge — Ann Dunkin

(fedscoop) The Department of Energy is working to increase the availability of datasets held by the agency for research, while also protecting sensitive data. Speaking Tuesday at the ACT-IAC 2021 Imagine Nation conference in Hershey, Penn., the department’s CIO Ann Dunkin outlined the dual challenge of fostering collaboration for research while also preserving national security. “We have so much data across the Department of Energy (DOE), and the challenge is to figure out how to move data between our labs and share it while still maintaining that propriety sense our labs each have about their data,” Dunkin said. Read More

Few-Shot Machine Learning Explained: Examples, Applications, Research

(iot for all) Data is what powers machine learning solutions. Quality datasets enable training models with the needed detection and classification accuracy, though sometimes the accumulation of sufficient and applicable training data that should be fed into the model is a complex challenge. For instance, to create data-intensive apps human annotators are required to label a huge number of samples, which results in complexity of management and high costs for businesses. In addition to that, there is the difficulty associated with data acquisition related to safety regulations, privacy, or ethical concerns. Read More

Artificial intelligence sheds light on how the brain processes language

(ScienceDaily) In the past few years, artificial intelligence models of language have become very good at certain tasks. Most notably, they excel at predicting the next word in a string of text; this technology helps search engines and texting apps predict the next word you are going to type. The most recent generation of predictive language models also appears to learn something about the underlying meaning of language. These models can not only predict the word that comes next, but also perform tasks that seem to require some degree of genuine understanding, such as question answering, document summarization, and story completion. Read More