New center tackles brain-inspired computing research

The center will support the research goals of the U.S. Air Force and Space Force.

By AMRUSHA BUDDHIRAJU
Joshua Yang
Professor Joshua Yang’s research aims to build more efficient computing systems to meet the demand for artificial intelligence and data processing. (Joshua Yang)

Joshua Yang, a professor of electrical and computer engineering, secured a five-year grant in November 2023 from the United States Air Force to establish and lead a Center of Excellence for researching neuromorphic computing. The COE program is set to launch June 2024 with offices in Seaver Science Center.

Neuromorphic computing is a process in which computers are designed to mimic the human brain in processing information and performing tasks. The center aims to address the U.S. Department of Defense’s research objective of building efficient computing devices that can withstand extreme environmental conditions often encountered in U.S. Air Force and Space Force applications.


Daily headlines, sent straight to your inbox.

Subscribe to our newsletter to keep up with the latest at and around USC.

“Neuromorphic computing is at the center of future machine-learning and artificial intelligence with improved efficiency,” Yang wrote in a statement to the Daily Trojan. “The field is a new way of computing that strives to imitate the human brain by combining computing power, resiliency, learning-efficiency and energy-efficiency.”

The grant opportunity, entitled “Center of Excellence: Extreme Neuromorphic Materials and Computing,” is a collaborative project among the Air Force Office of Scientific Research, Air Force Research Lab Technical Directorates and university researchers. Yang will lead as director of the Center of Neuromorphic Computing and Extreme Environment, along with other researchers from UCLA, Duke University, the Rochester Institute of Technology and the University of Texas at San Antonio.

According to an announcement released by AFRL, the demand for artificial intelligence and data processing computation doubles every three-and-a-half months — but the processors’ performance doubles every three-and-a-half years, emphasizing the need for robust and efficient computing systems. 

Yang’s primary research focuses on identifying alternatives to traditional materials and devices used in computing systems to enable efficient AI and machine learning capabilities. Yang is also co-director of the Institute for the Future of Computing at USC.

The DOD aims to build computing devices that can sustain harsh environments. These devices, used in unmanned aerial vehicles and satellite operations, are exposed to corrosion, erosion and extreme heat, which necessitate resilient systems for optimal performance and longevity. 

Tanvi Gandhi, a graduate student studying electrical and computer engineering with an emphasis in machine learning and data science, said the research could change computing and could open up new horizons.

“I’ve worked on various [machine learning and data science] projects,” Gandhi said. “It takes huge processing power and energy consumption to train pretty much any noteworthy neural network … so what could even be better by mimicking the way our brain functions [in computing systems]?”  

Janhavi Pradhan, a graduate student studying applied data science, said she was excited about the research because it hasn’t been explored much.

“A parallel that I can draw is how deep learning also mimics the way [the] human brain thinks,” Pradhan said. “That’s why it has gained popularity over regular [machine learning] models. So, processors that model [the] human brain could also possibly have better performance. That’s something that excites me about this.”

© University of Southern California/Daily Trojan. All rights reserved.