Center for Advanced Electronics through Machi

Center for Advanced Electronics through Machi

image: Elyse Rosenbaum, director of CAEML
View moreover

Credit: The Grainger College of Engineering, University of Illinois Urbana-Champaign

The Center for Advanced Electronics through Machine Learning (CAEML), funded as a Phase I IUCRC by the National Science Foundation since 2017, has just received funding from NSF to proceed with a second five-year phase. Phase II research will officially begin on August 1, 2022.

CAEML’s research vision is to apply machine learning to the design of optimized microelectronic circuits and systems, thereby increasing the efficiency of Electronic Design Automation (EDA) and resulting in a radically reduced design cycle time and reliability. improved.

In keeping with the nature of NSF’s IUCRC (Industrial-Academic Cooperative Research Center) program, CAEML focuses on partnerships with industry members who contribute to research funding and help guide the center’s research by identifying the industry’s major needs and acting as mentors for research projects. Administrative costs are covered by NSF and industry participants get IP rights and access to first-rate graduate students doing industry-relevant research.

According to CAEML Director Elyse Rosenbaum, “University attendees are gratified by the knowledge that their work will influence industry practices. Participation in the center is particularly beneficial to students and junior faculty, who learn about real-world technical challenges and can apply their creativity and knowledge to solutions. ”

Phase I results to date include publishing over 70 articles (some with industry co-authors), delivering 20 webinars, and receiving 11 awards or award nominations. Twelve students who participated in Phase I of CAEML have already been hired in full-time positions by the companies associated with CAEML. Furthermore, before graduation, many students did summer internships at CAEML companies.

In Phase II, the CAEML team will focus its attention on five technical challenges identified by the centre’s industry advisory committee: analog circuit design in advanced semiconductor technologies; end-to-end channel models for signal integrity analysis; security of IP, signals and design data; system templates that provide useful information to support the reliability of the system; and secure access to proprietary data for machine learning (ML), to facilitate research cooperation between companies and between companies and universities.

Rosenbaum says, “By leveraging Phase I research findings, CAEML Phase II is expected to make significant progress in design optimization and efficiency by developing the ability to learn inverse surrogate models, such as an inverse neural network, which provides a mapping from the specific design to suitable design outputs. The center will continue its work on the interconnections of the ML models at the component level, recognizing that the accuracy of the end-to-end system model is of utmost importance. CAEML will continue to demonstrate models that take into account the variability and uncertainty, which is a requirement for obtaining a high manufacturing yield “.

CAEML is a partnership between the University of Illinois Urbana-Champaign, North Carolina State University, and the Georgia Institute of Technology. In addition to Director Rosenbaum, who is Melvin’s Professor of Electrical and Computer Engineering and Anne Louise Hassebrock at UIUC, the leadership team is comprised of Paul Franzon, who is Cirrus Logic’s Distinguished Professor of Electrical and Computer Engineering and Director of ECE degree programs at NCSU and Madhavan Swaminathan, who is John Pippin Chair in Microsystems Packaging & Electromagnetics and Director of the Packaging Research Center at Georgia Tech.

The Phase I industry members of the center have different interests in microelectronic circuits and systems. The center will welcome old and new members in Phase II.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of press releases published on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.

Leave a Comment

Your email address will not be published.