Lighting the Way: The Revolutionary Shift to Optical AI Processors

Computer Memory Matrix Circuit Art Illustration

Nathan Youngblood, an assistant professor at the University of Pittsburgh, has been awarded grants from the NSF and AFOSR to further his research in optical computing and phase-change materials. His work aims to address the limitations of current computing hardware by developing more efficient, reliable, and fast optical computing systems. This research not only promises to enhance AI’s computational power but also to improve modern computing systems’ speed and efficiency. Credit:

University of Pittsburgh’s Nathan Youngblood is pioneering optical computing to boost AI and computing efficiency, supported by significant grants and focused on creating a diverse tech workforce.

The exponential demand for high computing power is far exceeding the capabilities of current electronic systems; however, engineers at the University of Pittsburgh are shining a light on new solutions.

Nathan Youngblood, principal investigator and assistant professor of electrical and computer engineering at Pitt’s Swanson School of Engineering, received a $552,166 Faculty Early Career Development Award from the National Science Foundation (NSF) and a $449,240 award from the Air Force Office of Scientific Research (AFOSR) through its Young Investigator Program (YIP) to continue his pioneering work in phase-change materials and optical computing.
“Dr. Youngblood is a rising star and one of the finest young researchers, scholars, and educators at Pitt Engineering,” said Alan George, Department Chair, R&H Mickle Endowed Chair, and Professor of Electrical and Computer Engineering and SHREC founder. “His two latest achievements – the CAREER Award and the AFOSR Young Investigator Award – are truly exceptional and we are so proud of him and excited about his growing research program and group of students.”

Optical computing, also called photonic computing, has shown promise over conventional hardware by using light waves produced by lasers or other sources for data storage, data processing, or data communication for computing. However, current technology limits its practicality.

With these awards, Youngblood will be investigating two different approaches to improve the speed, reliability, and efficiency of optical computing. The first approach focuses on using the wave-like nature of light to increase the efficiency of optical computing while the second focuses on improving optical memories to increase computational throughput.

Computing in the World of AI

For his CAREER Award, Youngblood’s will focus on developing high-efficiency optical computing hardware to address crucial challenges of artificial intelligence (AI).

“As AI applications services continue to become more prominent, we need the computing power to be able to support them,” Youngblood said. “There have been notable advancements in modern computers, but gains in traditional hardware efficiency are unable to keep pace with these data-hungry systems. Optical computing makes it possible.”

When current computing methods try to meet the demands of AI, unwanted heat is created because of the vast amounts of data moving at high speeds through the metal wires of the processor.

“Photons don’t have this heating issue, so you can process data much faster using light,” Youngblood explained. “Right now, however, optical processors aren’t powerful enough, accurate enough, or efficient enough to be truly useful for AI.”

Thanks to funding through Pitt’s Momentum Funds, Youngblood was able to secure an initial seeding grant and preliminary data for his CAREER Award.

“I’m incredibly thankful for Pitt’s help in jumpstarting this research,” Youngblood said.

An Upgrade in Modern Computing

It’s pretty clear modern computing systems have hit their limit.

Existing computer hardware is hindered by the movement of data between memory and processing cores, reducing computing speeds and creating unwanted heat in the machine.

Through the Young Investigator Program, Youngblood will create photonic hardware which enables computation to occur in the optical memory array itself, drastically reducing the movement of data. His lab will conduct research in three main thrusts: improving the efficiency, reliability, and repeatability of electrically programmable phase-change photonic memory; designing fully analog multilayer photonic networks for fast and efficient computing; and demonstrating a multi-layer, fully analog photonic in-memory accelerator on chip.

The outcomes of this work will advance the development of novel materials for reconfigurable photonic devices and integrate these components into optoelectronic computational systems.

“The resulting platform is expected to have significant impact for Air and Space Force applications requiring ultra-low latency computation, target discrimination, and autonomous navigation where there is an immediate need for extremely high speed information processing,” Youngblood said.

The project, “Photonic in-memory accelerators for low-latency and efficient computing,” is part of the $21.5 million given to YIP recipients who receive three-year grants of up to $450,000. Individuals selected must show exceptional ability and promise for conducting basic research of the Department of the Air Force relevance.

In addition to the scientific contributions to the next step in optical and modern computing, Youngblood’s CAREER award will also help him cultivate a diverse high-tech workforce in the greater Pittsburgh area. Initiatives include creating affordable educational tools exposing students to nanotechnology applications in AI, conducting STEM workshops in collaboration with Pitt’s outreach program (LEAD), and mentoring undergraduate researchers through Pitt’s EXCEL summer research program. Voluntary assessments will measure educational outcomes, providing quantifiable metrics for the project’s broader impact on workforce diversity and innovation in AI.

Be the first to comment on "Lighting the Way: The Revolutionary Shift to Optical AI Processors"

Leave a comment

Email address is optional. If provided, your email will not be published or shared.