The Intelligence Advanced Research Projects Activity (IARPA) is seeking “seedling” research topics for next-generation microelectronics in support of artificial intelligence (MicroE4AI).
In a broad agency announcement (BAA), IARPA announced that it is soliciting proposals for early-stage research, or seedlings, from “multidisciplinary teams who can co-design micro-electronics for AI technologies” that will assist with improving the hardware, software, algorithm, and architecture ecosystem for specific AI applications; and developing new science, materials, and processing that will improve the performance of microelectronics used to run AI applications.
“[IARPA] invests in high-risk/high-payoff research programs that have the potential to provide our nation with an overwhelming intelligence advantage,” IARPA said. “The increasing utilization of, and dependency on, [AI] and machine learning (ML) makes the development of faster, more energy efficient, and more resilient computing important to the future of the national security of the United States and the nation’s leadership in AI/ML.”
IARPA goes on to say that data storage, data movement, and data analysis place limits on the computer models’ ability to implement AI and ML and that “faster, more energy efficient and resilient computing requires that challenges including the physical limits on transistors, electrical interconnects, and memory elements be overcome.”
Preference will be given to proposals that show “research with the ability to revolutionize hardware-software integration from material properties to hardware design, to system architecture, to software implementation,” IARPA said.
White papers for the BAA are due no later than May 17, 2021 and proposals are due June 30, 2021.