期刊
SCIENCE ADVANCES
卷 8, 期 14, 页码 -出版社
AMER ASSOC ADVANCEMENT SCIENCE
DOI: 10.1126/sciadv.abm8537
关键词
-
资金
- Samsung Research Funding and Incubation Center of Samsung Electronics [SRFC-TA1903-05]
- National Research Foundation of Korea [NRF-2016M3D1A1027663, NRF-2019R1A2C2084114, NRF-2020M3F3A2A01081774]
In this study, integrated ferroelectric thin-film transistor (FeTFT) synaptic arrays were demonstrated to provide efficient parallel programming and data processing for CNNs through selective and accurate control of polarization in the ferroelectric layer.
Convolutional neural networks (CNNs) have gained much attention because they can provide superior complex image recognition through convolution operations. Convolution processes require repeated multiplication and accumulation operations, which are difficult tasks for conventional computing systems. Compute-in-memory (CIM) that uses parallel data processing is an ideal device structure for convolution operations. CIM based on two-terminal synaptic devices with a crossbar structure has been developed, but unwanted leakage current paths and the high-power consumption remain as the challenges. Here, we demonstrate integrated ferroelectric thin-film transistor (FeTFT) synaptic arrays that can provide efficient parallel programming and data processing for CNNs by the selective and accurate control of polarization in the ferroelectric layer. In addition, three-terminal FeTFTs can act as both nonvolatile memory and access device, which tackle issues from two-terminal devices. An integrated FeTFT synaptic array with parallel programming capabilities can perform convolution operations to extract image features with a high-recognition accuracy.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据