Programming memristor arrays with arbitrarily high precision for analog computing
成果类型：
Article
署名作者：
Song, Wenhao; Rao, Mingyi; Li, Yunning; Li, Can; Zhuo, Ye; Cai, Fuxi; Wu, Mingche; Yin, Wenbo; Li, Zongze; Wei, Qiang; Lee, Sangsoo; Zhu, Hengfang; Gong, Lei; Barnell, Mark; Wu, Qing; Beerel, Peter A.; Chen, Mike Shuo-Wei; Ge, Ning; Hu, Miao; Xia, Qiangfei; Yang, J. Joshua
署名单位：
University of Southern California; University of Massachusetts System; University of Massachusetts Amherst
刊物名称：
SCIENCE
ISSN/ISSBN：
0036-13262
DOI：
10.1126/science.adi9405
发表日期：
2024-02-23
页码：
903-910
关键词：
摘要：
In-memory computing represents an effective method for modeling complex physical systems that are typically challenging for conventional computing architectures but has been hindered by issues such as reading noise and writing variability that restrict scalability, accuracy, and precision in high-performance computations. We propose and demonstrate a circuit architecture and programming protocol that converts the analog computing result to digital at the last step and enables low-precision analog devices to perform high-precision computing. We use a weighted sum of multiple devices to represent one number, in which subsequently programmed devices are used to compensate for preceding programming errors. With a memristor system-on-chip, we experimentally demonstrate high-precision solutions for multiple scientific computing tasks while maintaining a substantial power efficiency advantage over conventional digital approaches.