Sandia National Labs Academic Programs Collaboration Report

The Center for Advanced Electronics through Machine Learning leads cooperative research

Industry-University Cooperative Research Centers such as the Center for Advanced Electronics Through Machine Learning (CAEML) are making tremendous strides applying machine learning (ML) to advanced electronics through collaborative, focused research. Illinois is leading CAEML under their Coordinated Science Laboratory, and their mission—enabling fast, accurate design and verification of microelectronic circuits and systems through machine learning— aligns with Sandia’s goal to facilitate rapid development and deployment of scalable, rad aware, low-cost modeling hierarchy at all levels of nuclear deterrence system integration. Biliana Paskaleva, a researcher in Sandia’s Component & Systems Analysis department, is leading Sandia’s engagement with CAEML. As a member of CAEML, Sandia works with the industry and academic partners to inform and help direct research goals and projects in areas relevant to the mission and the nation. These focus areas include: • Hardware security and trusted microelectronics • Accelerated design and qualification through rapid mod/sim • Microelectronics aging and reliability • AI/ML model credibility Illinois, Georgia Tech and North Carolina State University, all university partners of Sandia, are all achieving significant results in partnership through CAEML. Joshua Hanson, a PhD student in electrical engineering at Illinois and a contributing

CAEML industry member, was recruited as a summer intern for Sandia’s Computer Science Research Institute and is now a year-round intern working on the application of machine learning techniques for the development of data-driven radiation-aware models for electrical devices and circuits. Hanson has already co authored and published several journal papers, with one featured in the February 2022 issue of IEEE Transactions on Nuclear Science. In April 2022, the National Science Foundation (NSF) published an article on a research achievement in semiconductor packaging accomplished by CAEML member Georgia Tech. Noteworthy results in ML hardware solutions achieved with CAEML member North Carolina State were also highlighted in the October 2022 issue of DesignNews. CAEML received a Phase II award from the NSF to continue driving innovation through cooperative fundamental research for the next 5 years in science, engineering, and technology on behalf of the nation. CAEML Director Elyse Rosenbaum from Illinois said, “Leveraging research results from Phase I, Phase II CAEML is expected to make significant advances in design optimization and efficiency.”

The Center for Advanced Electronics Through Machine Learning

is a cooperative research center.

63

2021-2022 Collaboration Report