J. Joshua Yang, Guest Editor
I am J. Joshua Yang, a professor in the Electrical and Computer Engineering Department at the Viterbi School of Engineering, University of Southern California. My research focus has been on emerging computing devices, and I also serve as the founding chair of the IEEE Electron Devices Society (EDS) Neuromorphics Technical Committee. I am thrilled to introduce a new special issue of the IEEE Electron Devices Magazine (ED-M) dedicated to neuromorphic computing.
Neuromorphic computing has garnered significant attention in recent years for two main reasons. First, within computer engineering, there is a growing need to revolutionize our existing computing systems based on von Neumann architectures and traditional CMOS devices. As the von Neumann bottleneck becomes more pronounced in data-centric tasks like machine learning and artificial intelligence and as Moore’s law for CMOS scaling reaches its limits, the quest for new computing paradigms becomes increasingly urgent. The human brain, being an incredibly efficient computing system, provides valuable insights for computer engineers. Neuromorphic computing, inspired by the brain’s computing and learning processes, emerges as a promising avenue. Second, in the field of neuroscience, we have embarked on a prolonged journey to understand the brain, yet with limited success. In addition to direct research on biotissues (often akin to a black box) neuromorphic systems constructed using artificial electronic devices, capable of faithfully emulating biological components (such as synapses, neurons, and dendrites), offer a transparent framework. Such systems not only enable measurement of input–output relationships but also provide insights into the functioning of every individual node within the system at any required moment.
The realization of neuromorphic systems hinges on developing neuromorphic devices as fundamental building blocks. This responsibility falls to the electronic device community, and it is precisely this focus that defines the purpose of this special issue. Four preeminent research groups from around the globe were invited to contribute to this issue, covering various aspects of neuromorphic devices. Prof. Leon Chua of the University of California, Berkeley, the inventor of the memristor and a celebrated member of EDS, discusses the creation of biomimetic neurons using homemade US$10 Chua corsage memristors in his featured article [A1]. Prof. Shimeng Yu and his team from the Georgia Institute of Technology provide a summary of recent advancements in the development of artificial synapses using capacitive devices, as opposed to the more commonly used resistive devices, in their invited article [A2]. Dr. James B. Aimone and his colleagues from Sandia National Laboratories share intriguing insights into the compatibility of stochastic devices with other components in neuromorphic hardware. Their invited article [A3] delves into this important aspect. Finally, Prof. Bin Gao, Prof. Huaqiang Wu, and their team from Tsinghua University present a comprehensive review of the latest developments in neuromorphic chips based on emerging memory devices in their invited article [A4].
In conclusion, as the guest editor of this special issue, I would like to express my gratitude for the generous support of Dr. Joachim N. Burghartz, the editor in chief of ED-M, and Dr. John Dallesasse, vice president of the IEEE EDS Technical Committees. I am also immensely thankful to all the authors who have shared their profound insights in this stimulating special issue on neuromorphic computing in ED-M. We also acknowledge TetraMem Inc. for providing the cover image of the special issue.
J. Joshua Yang (jjoshuay@usc.edu) is a professor in the Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA 90089 USA. His current research interest is post-CMOS hardware for neuromorphic computing, machine learning, and artificial intelligence. He is the founding chair of the IEEE Neuromorphic Computing Technical Committee. He serves as an associate editor of Science Advances (AAAS). He is an IEEE Fellow and a National Academy of Inventors Fellow.
[A1] L. O. Chua, “Homemade US$ 10 Chua corsage memristor: Use it to make the poor man’s biomimetic neurons,” IEEE Electron Devices Mag., vol. 1, no. 2, pp. 2-14, Sep. 2023, doi: 10.1109/ MED.2023.3296483.
[A2] S. Yu, Y.-C. Luo, T.-H. Kim, and O. Phadke, “Nonvolatile capacitive synapse: Device candidates for charge domain compute-in-memory,” IEEE Electron Devices Mag., vol. 1, no. 2, pp. 25-34, Sep. 2023, doi: 10.1109/MED.2023.3293060.
[A3] J. B. Aimone and S. Misra, “Will stochastic devices play nice with others in neuromorphic hardware? There’s more to a probabilistic system than noisy devices,” IEEE Electron Devices Mag., vol. 1, no. 2, pp. 52-58, Sep. 2023, doi: 10.1109/ MED.2023.3298873.
[A4] Q. Wei, B. Gao, J. Tang, H. Qian, and H. Wu, “Emerging memory-based chip development for neuromorphic computing: Status, challenges, and perspectives,” IEEE Electron Devices Mag., vol. 1, no. 2, pp. 35-51, Sep. 2023, doi: 10.1109/MED.2023.3296084.
Digital Object Identifier 10.1109/MED.2023.3307978
Date of current version: 15 September 2023