返回
2D Materials for Neuron Devices and Neuromorphic Computing
DOI:10.1002/smll.202506454.png)
摘要
En 中文
With the rapid development of artificial general intelligence and the energy-efficiency limitations of traditional architectures, bio-inspired neuromorphic computing systems based on brain-like learning offer a promising pathway. Compared to conventional silicon-based devices, 2D materials garner significant attention due to their atomic-scale thickness, tunable optoelectronic properties, and high degree of freedom in heterostructure integration. These exceptional physical characteristics establish 2D materials as strong contenders for neuromorphic hardware. This review systematically introduces 2D material-based artificial neuron devices, summarized across four categories: memristive-type, transistor-type, reconfigurable-type, and optoelectronic-type devices. Next, a development roadmap for biologically inspired neuromorphic systems is summarized, drawing insights from the human brain's learning pathways. Finally, the review discusses future opportunities and challenges for 2D material neuromorphic systems. The evidence indicates that 2D material-based neuromorphic computing systems represent a potential and viable route for future advancements.
Keyword:
2D material
artificial neuron device
in-memory computing
neuromorphic computing
期刊
IF:
12.1
论文数:
3.0W
被引数:
16.4W
机构
引用论文
Breakdown of accommodation in nerve: a possible role for persistent sodium current神经中的传导阻滞:持续钠电流可能的作用
Two-dimensional semiconductor transistors and integrated circuits for advanced technology nodes
NATIONAL SCIENCE REVIEW
IF17.1
Progress and Challenges for Memtransistors in Neuromorphic Circuits and Systems神经形态电路和系统中记忆晶体管的进展和挑战
ADVANCED MATERIALS
IF26.8
An index-free sparse neural network using two-dimensional semiconductor ferroelectric field-effect transistors一种无需索引的稀疏神经网络,采用二维半导体铁电场效应晶体管
NATURE ELECTRONICS
IF40.9
A charge-density-wave oscillator based on an integrated tantalum disulfide-boron nitride-graphene device operating at room temperature基于集成的二硫化钽-氮化硼-石墨烯器件的室温电荷密度波振荡器
NATURE NANOTECHNOLOGY
IF34.9
Quasi-Hodgkin-Huxley Neurons with Leaky Integrate-and-Fire Functions Physically Realized with Memristive Devices准hodgkin-huxley神经元,具有通过忆阻器件物理实现的泄漏集成和发射功能
ADVANCED MATERIALS
IF26.8

