As a part of an ongoing effort to maintain you knowledgeable about our newest work, this weblog put up summarizes some current publications from the SEI within the areas of machine studying, moral synthetic intelligence, cloud adoption, and mixed-criticality methods. These publications spotlight the newest work of SEI technologists in these areas.
We’ve additionally included hyperlinks to a just lately revealed e-book and accompanying podcast that’s the results of a multi-year analysis and growth imaginative and prescient and roadmap for engineering next-generation software-reliant methods.
This put up features a itemizing of every publication, writer(s), and hyperlinks the place they are often accessed on the SEI web site.
Architecting the Way forward for Software program Engineering: A Nationwide Agenda for Software program Engineering Analysis & Improvement
by Anita Carleton, Mark H. Klein, John E. Robert, Erin Harper, Robert Ok Cunningham, Dionisio de Niz, John T. Foreman, John B. Goodenough, James D. Herbsleb, Ipek Ozkaya, Douglas Schmidt (Vanderbilt College), Forrest Shull
Software program is significant to our nation’s international competitiveness, innovation, and nationwide safety. It additionally ensures our trendy lifestyle and permits continued advances in protection, infrastructure, healthcare, commerce, training, and leisure. As a part of its work as a federally funded analysis and growth middle (FFRDC) targeted on utilized analysis to enhance the follow of software program engineering, the Carnegie Mellon College Software program Engineering Institute led the group in creating this multi-year analysis and growth imaginative and prescient and roadmap for engineering next-generation software-reliant methods.
An government abstract can also be accessible.
Hearken to/view the podcast by Anita Carleton and Forrest Shull detailing this initiative.
Utilizing Machine Studying to Enhance NPC Constancy
by Dustin D. Updyke, Thomas G. Podnar, Geoffrey B. Dobson, John Yarger
Experiences that appear actual to gamers in coaching and train situations improve studying. Enhancing the constancy of automated non-player characters (NPCs) can improve the extent of realism felt by gamers. On this report, we describe how we used machine studying (ML) modeling to create decision-making preferences for NPCs. In our analysis, we check ML options and make sure that NPCs can exhibit lifelike laptop exercise that improves over time.
Implementing the DoD’s Moral AI Rules
by Alexandrea Van Deusen and Carol J. Smith
Alexandrea Van Deusen and Carol Smith, each with the SEI’s AI Division, talk about a current undertaking by which they helped the Protection Innovation Unit (DIU) of the U.S. Division of Protection to develop tips for the accountable use of synthetic intelligence (AI), primarily based on the DoD’s Moral Rules for AI. These tips can function a information for organizations in trade and authorities to implement accountable AI issues into follow in real-world applications.
View/hearken to the podcast.
A Prototype Set of Cloud Adoption Threat Components
by Christopher J. Alberts
This report presents the outcomes of a examine that the SEI carried out to determine a prototype set of danger components for the adoption of cloud applied sciences. These danger components cowl a broad vary of potential issues that may have an effect on a cloud initiative, together with enterprise technique and processes, expertise administration and implementation, and organizational tradition.
The publication of this report is an preliminary step within the growth of cloud adoption danger components relatively than the fruits of SEI work on this space. This report identifies a variety of potential future growth and transition duties associated to the MRD for cloud adoption. The SEI MRD methodology defines a time-efficient, mission-oriented strategy for assessing danger in mission threads, enterprise processes, and organizational initiatives.
Obtain the white paper.
Cloud Safety Greatest Practices Derived from Mission Thread Evaluation
by Timothy Morrow, Vincent LaPiana, Donald Faatz, Angel Luis Hueca, Nathaniel Richmond
This report presents 4 essential safety practices which are sensible and efficient for enhancing the cybersecurity posture of cloud-deployed data expertise (IT) methods. These practices assist to handle the dangers, threats, and vulnerabilities that organizations face in deploying or shifting purposes and methods to a cloud service supplier (CSP).
The practices tackle cloud safety points that customers are experiencing, illustrated by a number of current cloud safety incidents. The report demonstrates the practices by means of examples utilizing cloud companies accessible from Amazon Net Service (AWS), Microsoft, and Google.
The offered practices are geared towards small and medium-sized organizations; nonetheless, all organizations, impartial of measurement, can use these practices to enhance the safety of their cloud utilization. The main target right here is on hybrid deployments the place some IT purposes deploy or transfer to a CSP whereas different IT purposes stay within the group’s knowledge middle. Small and medium-sized organizations doubtless have restricted assets; the place doable, these practices describe implementation approaches which may be efficient in limited-resource conditions.
This report is a 2021 replace of a previous report.
Obtain the SEI technical report.
Evolvable Technical Reference Frameworks for Blended-Criticality Programs
by Nickolas Guertin and Douglas Schmidt
Nickolas Guertin and Douglas Schmidt discuss with Suzanne Miller about methods for creating architectures for large-scale, advanced methods that comprise features with a variety of necessities. This is likely one of the most difficult areas in U.S. Division of Protection acquisition, and this strategy and the methods mentioned are essential to the way forward for our massive methods.
Hearken to/view the podcast.