• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Advanced functional materials >A Review of Artificial Spiking Neuron Devices for Neural Processing and Sensing
【24h】

A Review of Artificial Spiking Neuron Devices for Neural Processing and Sensing

机译：A Review of Artificial Spiking Neuron Devices for Neural Processing and Sensing

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相关主题

摘要

A spiking neural network (SNN) inspired by the structure and principles ofthe human brain can significantly enhance the energy efficiency of artificialintelligence computing by overcoming the bottlenecks of the conventionalvon Neumann architecture with its massive parallelism and spike transmissions.The construction of artificial neurons is important for the hardwareimplementation of an SNN, which generates spike signals when enoughsynaptic signals are gathered. Because circuit-level artificial neurons withcomparator and reset circuits require considerable hardware area, intensiveefforts are devoted in recent years for building artificial neurons at thedevice level for better area efficiency. Furthermore, artificial sensory neurondevices, which perform neural processing and sensing concurrently, haverecently been developed in order to reduce the hardware cost and energyconsumption of traditional sensory systems through in-sensor computing.This review article surveys and benchmarks the recent progress of artificialneuron devices for neural processing and sensing. First, various artificialneuron devices are summarized, including single-transistor neurons (1T-neurons),memristor neurons, phase-change neurons, magnetic neurons, andferroelectric neurons. Next, cointegration technologies with artificial synapticdevices and artificial sensory neurons for in-sensor computing are introduced.Finally, the challenges and prospects for developing artificial neurondevices are discussed.

著录项

获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号