Dr. Peng Li, professor in the Department of Electrical and Computer Engineering at Texas A&M University, has received a grant from the National Science Foundation (NSF) and Semiconductor Research Corporation (SRC) to help find revolutionary solutions that will enable more energy-efficient computing.
Li’s work is aimed at attaining brain-like learning performance by imitating how the brain represents, processes and learns from information. More specifically, by developing models of computation based on the third-generation spiking neural networks and efficient adaptive processor architectures.
“Drawing inspiration from the human brain offers rich insights and opportunities for developing disruptive computing solutions desperately needed in today’s machine learning and big data applications,” Li said. “It will enable the development of novel information processing paradigms by mimicking key computational characteristics of the brain.”
Li’s project will develop spike-dependent learning mechanisms to allow training of complex recurrent spiking neural networks. Self-adaptive processor architectures with integrated on-chip learning, light-weight runtime learning performance prediction and energy management will be developed to maximize the energy efficiency of the proposed neural processors while providing a guarantee for learning performance.
Li received nearly $500,000 for his project that was one of nine three-year projects awarded a total of $21.6 million by the NSF and SRC. The topics to be studied range from advanced brain-inspired computer architectures like Li’s to hybrid digital-analog designs. The goal of the program is to create a new type of computer that can proactively interpret and learn from data, solve unfamiliar problems using what it has learned and operate closer to the efficiency of the human brain.
“Only disruptive breakthroughs can enable computers to perform as the human brain does, in terms of problem-solving capability and lower power, which, for the human brain, is less than a light bulb’s worth of consumption,” said Dimitris Pavlidis, NSF’s Directorate for Engineering program director for the Energy-Efficient Computing: from Devices to Architectures (E2CDA) initiative.
Computing capabilities in the United States rely on the continuous research and development of new computing systems with rapidly increasing performance. However, improvements in computing performance are severely limited by the amount of energy needed.
According to Pavlidis, the three-year projects consider simultaneously novel approaches — including developing nanoscale devices and materials and integrating them into three-dimensional systems — while inventing new computer architectures to process, store and communicate data.
The initiative to create new types of high efficiency computing systems was detailed by the White House when it announced the Nanotechnology-Inspired Grand Challenge for Future Computing. This effort also aligns with one of the NSF’s big ideas for future investment as well as the National Strategic Computing Initiative to advance the “Work at the Human-Technology Frontier: Shaping the Future.”
To learn more about the projects visit the NSF’s website.