Sandia National Labs Academic Programs Collaboration Report

A new computing paradigm for solving complex probability problems The Co-designed Improved Neural Foundations Leveraging Inherent Physics Stochasticity (COINFLIPS) microelectronics co-design program, led by Sandia PI Brad Aimone, will receive $6 million over the next 3 years. The research, done in conjunction with Oak Ridge National Laboratory, New York University, the UT Austin, and Temple University in Philadelphia, will focus on developing “graphene-based synaptic transistors” that help create computers that learn and think like a human brain. COINFLIPS is one of only 10 national laboratory-led projects selected for funding by DOE’s Office of Science. “ What we want to do in this project is to leverage randomness. Instead of fighting it, we want to use it. ” said Aimone, who has led Sandia LDRD and Grand Challenge projects in the field. The funding award by DOE’s Office of Science for the 2021 COINFLIPS microelectronics co-design research program is a direct result of the successful workshop on AI-Enhanced Co-Design for Next Generation Microelectronics that Sandia hosted in April 2021. Over 40 individuals from all institutions attended the COINFLIPS kickoff meeting in November 2021.

CONTRIBUTOR Spotlight

Jean Anne Incorvia, UT Austin assistant professor, is a faculty collaborator working on the COINFLIPS collaboration. She observed, “The COINFLIPS project is a true co-design effort combining a broad range of expertise. It has been a great opportunity to contribute my research group’s expertise in unconventional computing with magnetic materials and devices, and to learn from others.”

Conventional computers can look at the optical illusion on the left and normally only see a vase or two faces. Sandia is laying the groundwork for a computer that, like our brains, can glance many times and see both.

77

2021-2022 Collaboration Report