Voltage-controlled magnetoelectric devices for neuromorphic diffusion process.

Journal: Nature communications
Published Date: May 30, 2025

Abstract

Neuromorphic diffusion models have become one of the major breakthroughs in the field of generative artificial intelligence. Unlike discriminative models that have been well developed to tackle classification or regression tasks, diffusion models aim at creating content based upon contexts learned. However, the more complex algorithms of these models result in high computational costs using today's technologies. Here, we develop a spintronic voltage-controlled magnetoelectric memory hardware for the neuromorphic diffusion process. The in-memory computing capability of our spintronic devices goes beyond current Von Neumann architecture, where memory and computing units are separated. Together with the non-volatility of magnetic memory, we can achieve high-speed and low-cost computing, which is desirable for the increasing scale of generative models in the current era. We experimentally demonstrate that the hardware-based true random diffusion process can be implemented for image generation and achieve comparable image quality to software-based training as measured by the Fréchet inception distance (FID) score, achieving ~10 better energy-per-bit-per-area over traditional hardware.

Authors

  • Yang Cheng
    Ministry of Agriculture Key Laboratory of Molecular Biology of Crop Pathogens and Insects, Institute of Insect Sciences, Zhejiang University, Hangzhou 310058, China.
  • Qingyuan Shu
    Department of Electrical and Computer Engineering, University of California, Los Angeles, CA, USA.
  • Albert Lee
    JC School of Public Health and Primary Care, The Chinese University of Hong Kong, Shatin, China.
  • Haoran He
    Department of Electrical and Computer Engineering, University of California, Los Angeles, CA, USA.
  • Ivy Zhu
    Department of Physics, The Ohio State University, Columbus, OH, USA.
  • Minzhang Chen
    Department of Electrical and Computer Engineering, University of California, Los Angeles, CA, USA.
  • Renhe Chen
    Department of Electrical and Computer Engineering, University of California, San Diego, CA, USA.
  • Zirui Wang
  • Hantao Zhang
    Department of Physics and Astronomy, University of California, Riverside, CA, USA.
  • Chih-Yao Wang
    Industrial Technology Research Institute, Taipei, Taiwan, ROC.
  • Shan-Yi Yang
    Industrial Technology Research Institute, Taipei, Taiwan, ROC.
  • Yu-Chen Hsin
    Industrial Technology Research Institute, Taipei, Taiwan, ROC.
  • Cheng-Yi Shih
    Industrial Technology Research Institute, Taipei, Taiwan, ROC.
  • Hsin-Han Lee
    Industrial Technology Research Institute, Taipei, Taiwan, ROC.
  • Ran Cheng
    Shenzhen Key Laboratory of Computational Intelligence, University Key Laboratory of Evolving Intelligent Systems of Guangdong Province, Department of Computer Science and Engineering, Southern University of Science and Technology, Shenzhen, 518055, China. Electronic address: ranchengcn@gmail.com.
  • Kang L Wang
    Department of Electrical and Computer Engineering, University of California, Los Angeles, CA, USA. wang@ee.ucla.edu.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals