检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Li, Yi Wang, Songqi Zhao, Yaping Wang, Shaocong Zhang, Woyu He, Yangu Lin, Ning Cui, Binbin Chen, Xi Zhang, Shiming Jiang, Hao Lin, Peng Zhang, Xumeng Qi, Xiaojuan Wang, Zhongrui Xu, Xiaoxin Shang, Dashan Liu, Qi Cheng, Kwang-Ting Liu, Ming Department of Electrical and Electronic Engineering The University of Hong Kong Hong Kong Laboratory of Microelectronic Devices & Integrated Technology Institute of Microelectronics Chinese Academy of Sciences Beijing100029 China ACCESS – AI Chip Center for Emerging Smart Systems InnoHK Centers Hong Kong Science Park Hong Kong State Key Lab of Fabrication Technologies for Integrated Circuits Institute of Microelectronics Chinese Academy of Sciences Beijing100029 China Frontier Institute of Chip and System Fudan University Shanghai200433 China College of Computer Science and Technology Zhejiang University Zhejiang310027 China Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Hong Kong University of Chinese Academy of Sciences Beijing100049 China Institute of Mind The University of Hong Kong Hong Kong

The rapid advancement of artificial intelligence (AI) has been marked by the large language models exhibiting human-like intelligence. However, these models also present unprecedented challenges to energy consumption and environmental sustainability, which are exacerbated by the increasing number of users and the development of even larger models. One promising solution is to revisit analogue computing, a technique that predates digital computing and exploits emerging analogue electronic devices, such as resistive memory, which features in-memory computing, high scalability, and nonvolatility that addresses the von Neumann bottleneck, slowdown of Moore’s law, and volatile DRAM of conventional digital hardware. However, analogue computing still faces the same challenges as before: programming nonidealities and expensive programming due to the underlying devices physics. Therefore, leveraging the efficiency advantage while mitigating the programming disadvantage of analogue computing with resistive memory is a major open problem in AI hardware and electronics communities. Here, we report a universal solution, software-hardware co-design using structural plasticity-inspired edge pruning to optimize the topology of a randomly weighted analogue resistive memory neural network. Software-wise, the topology of a randomly weighted neural network is optimized by pruning connections rather than precisely tuning resistive memory weights. Hardware-wise, we reveal the physical origin of the programming stochasticity using transmission electron microscopy, which is leveraged for large-scale and low-cost implementation of an overparameterized random neural network containing high-performance sub-networks. We implemented the co-design on a 40nm 256K resistive memory macro, observing 17.3% and 19.9% accuracy improvements in image and audio classification on FashionMNIST and Spoken digits datasets, as well as 9.8% (2%) improvement in PR (ROC) in image segmentation on DRIVE datasets, r

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Aligned monolayer MoS2 ribbons growth on sapphire substrate via NaOH-assisted chemical vapor deposition

引用

Science China Materials 2020年第6期63卷 1065-1075页

作者： Shike Hu Jing Li Xiaoyi Zhan Shuang Wang Longbiao Lei Yijian Liang He Kang Yanhui Zhang Zhiying Chen Yanping Sui Da Jiang Guanghui Yu Songang Peng Zhi Jin Xinyu Liu State Key Laboratory of Functional Materials for Informatics Shanghai Institute of Microsystem and Information TechnologyChinese Academy of SciencesShanghai 200050China Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of SciencesBeijing 100049China Microwave Devices and Integrated Circuits Department Institute of MicroelectronicsChinese Academy of SciencesBeijing 100029China

This study reports the growth of aligned monolayer molybdenum disulfide(MoS2)ribbons on a sapphire substrate via NaOH-assisted chemical vapor *** length of MoS2 ribbon is up to 400μ*** MoS2 ribbon has excellent single crystal properties,a carrier mobility of^150 cm^2V^-1s^-1,and a optical response of 103 mA W^-1 at a wavelength of 550 *** growth model of MoS2 ribbons was *** reacts with MoO3 to form sodium molybdate droplets,which increase the fluidity of the molybdenum source on the substrate,realizing the vapor-liquid-solid growth of MoS2 on *** monolayer MoS2 ribbons have two kinds of arrangements on the sapphire substrate,one is the oriented growth affected by the interlayer van der Waals force and the lattice,and the other is the aligned growth constrained by the sapphire *** results promote the basic research and device applications of MoS2,and introduce a new way of synthesizing other one dimensional(1D)and 2D nanostructures.

关键词： MoS2 aligned ribbons vapor-liquid-solid step epitaxy

来源：评论

学校读者我要写书评

暂无评论

Modeling IC Snapback Characteristics Using a VCCS Model for Circuit-Level ESD Simulation

Modeling IC Snapback Characteristics Using a VCCS Model for ...

引用

International Symposium on Physical & Failure Analysis of Integrated circuits

作者： Yunhao Li Yize Wang Yuan Wang Key Laboratory of Microelectronics Device and Circuits (MoE) Institute of Microelectronics Peking University Beijing China

Snapback behaviors under electrostatic discharge (ESD) stress are difficult to be simulated due to the limitation of SPICE tools. This paper proposes a novel way to model avalanche effect using a voltage-controlled current source (VCCS). This method greatly simplifies the calculation of the avalanche multiplication factor M and improving the convergence properties. This can change the exponential relationship of the M into a linear relationship of the gain of the VCCS. Meanwhile, this method can be used for mixed simulation at the circuit-level and device-level and model double snapbacks phenomenon appearing in power clamp circuits. Measurement and simulation of the new approach indicates that this approach is useful.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Oscillation neuron based on threshold switching characteristics of niobium oxide films

Oscillation neuron based on threshold switching characterist...

引用

Future Computing (IWOFC,2019 IEEE International Workshop on

作者： Qingxi Duan Zhaokun Jing Ke Yang Ru Huang Yuchao Yang Key Laboratory of Microelectronic Devices and Circuits (MOE) Peking University Beijing China

ISBN: (数字)9781728125336

ISBN: (纸本)9781728125343

Here we report threshold switching characteristics in niobium oxide films, which is in turn used to build an oscillation neuron. In particular, we show that strong correlation exists between the oxygen flow ratio during NbOx film deposition and the electrical properties of NbOx devices. The NbOx devices can show metallic, volatile threshold switching, and non-volatile resistive switching characteristics, depending on the oxygen flow ratio. The NbOx device with threshold switching can be used to form an oscillation neuron displaying three critical features: all-or-nothing oscillation, threshold-driven oscillation and input-modulated frequency response, which could be promising for applications as artificial neurons in hardware neural networks.

关键词： NbOx threshold switching oscillation neuron specific emission Oscillations switching characteristics OXIDE FILMS Neurons niobium oxides niobium Threshold frequency response

来源：评论

学校读者我要写书评

暂无评论

10-nm Channel Length Indium-Tin-Oxide transistors with Ion = 1860 μA/μm, Gm = 1050 μS/μm at Vds = 1 V with BEOL Compatibility

10-nm Channel Length Indium-Tin-Oxide transistors with Ion =...

引用

International Electron devices Meeting (IEDM)

作者： Shengman Li Chengru Gu Xuefei Li Ru Huang Yanqing Wu Institute of Microelectronics and Key Laboratory of Microelectronic Devices and Circuits (MoE) Peking University Beijing Wuhan National High Magnetic Field Center and School of Optical and Electronic Information Huazhong University of Science and Technology Wuhan China Frontiers Science Center for Nano-optoelectronics Peking University Beijing China

ISBN: (数字)9781728188881

ISBN: (纸本)9781728188898

In this paper, we successfully realized a shortest 10-nm channel length transistor based on ultrathin 3.5-nm indium tin oxide channel. Using 5-nm lanthanum-doped hafnium oxide (HfLaO) high- κ dielectric, the 10- nm channel ultrathin device architecture with wide bandgap exhibits excellent switching behavior with on/off ratio exceeding 10 10 and ultra-low leakage current of 40 fA/μm. Record-high on-state current of 1860 μA/μm and transconductance (g m ) of more than 1000 μS/μm have been achieved, benefited from a rather small contact resistance of 162 Ω·μm. R on of the 10-nm ITO transistor was smaller than 500 Ω·μm. Adopting the MIT Virtual Source (MVS) model, we extracted the saturation injection velocity ν x0 up to 8.8 × 10 6 cm/s and the resulting mean free path λ mfp of ITO transistor is 21.6 nm. Radio frequency transistor with 30-nm-long channel exhibited record high radio-frequency (RF) performance with small-signal current gain (f T ) of 20 GHz and maximum oscillation frequency (f max ) of 13 GHz. These two metrics contribute to √(f T × f max ) exceeding 15 GHz, confirming overwhelming superiority compared to RF transistors based on other ultrathin novel channel materials such as MoS 2 , BP and metal-oxides. Finally, we employed bootstrapped mode (BST) inverters to fabricate a 5-stage ring oscillator, and achieved the record-low propagation delay of 0.4 ns/stage among metal-oxides.

关键词： Ring oscillators Radio frequency Indium tin oxide Frequency measurement Transistors Sulfur Molybdenum

来源：评论

学校读者我要写书评

暂无评论

MXenes induce epitaxial growth of size-controlled noble nanometals:A case study for surface enhanced Raman scattering(SERS)

引用

Journal of Materials Science & Technology 2020年第5期40卷 119-127页

作者： Renfei Cheng Tao Hu Minmin Hu Changji Li Yan Liang Zuohua Wang Hui Zhang Muchan Li Hailong Wang Hongxia Lu Yunyi Fu Hongwang Zhang Quan-Hong Yang Xiaohui Wang Shenyang National Laboratory for Materials Science Institute of Metal ResearchChinese Academy of SciencesShenyang 110016China School of Materials Science and Engineering University of Science and Technology of ChinaShenyang 110016China University of Chinese Academy of Sciences Beijing 100049China National Engineering Research Center for Equipment and Technology of Cold Strip Rolling College of Mechanical EngineeringYanshan UniversityQinhuangdao 066004China Department of Materials Science and Engineering Monash UniversityClaytonVictoria 3800Australia Institute of Microelectronics Key Laboratory of Microelectronic Devices and CircuitsPeking UniversityBeijing 100871China School of Materials Science and Engineering Zhengzhou UniversityZhengzhou 450001China School of Chemical Engineering&Technology Tianjin UniversityTianjin 300072China

Noble nanometals are of significance in both scientific interest and technological applications,which are usually obtained by conventional wet-chemical *** surfactants are always used in the synthesis to prevent unexpected overgrowth and aggregation of noble ***,the surfactants are hard to remove and may interfere with plasmonic and catalytic studies,remaining surfactant-free synthesis of noble nanometals a ***,we report an approach to epitaxial growth of sizecontrolled noble nanometals on *** piloted by density functional theory calculations,along with work function experimental determination,kinetic and spectroscopic studies,epitaxial growth of noble nanometals is initiated via a mechanism that involves an in situ redox *** the redox,MXenes as two-dimensional solid reductants whose work functions are compatible with the reduction potentials of noble metal cations,enable spontaneous donation of electrons from the MXenes to noble metal cations and reduce the cations into nanoscale metallic metals on the outmost surface of *** surfactants nor external reductants are used during the whole synthesis process,which addresses a long-standing interference issue of surfactant and external reductant in the conventional wet-chemical ***,the MXenes induced noble nanometals are ***,noble nanometals firmly anchored on MXenes exhibit excellent performance towards surface enhanced Raman *** developed strategy will promote the nanostructure-controlled synthesis of noble nanometals,offering new opportunities to further improve advanced functional properties towards practical applications.

关键词： Two-dimensional materials MXene In situ redox Noble metal SERS

来源：评论

学校读者我要写书评

暂无评论

An Automatic Slope-Calibrated Ramp Generator for Single-Slope ADCs

An Automatic Slope-Calibrated Ramp Generator for Single-Slop...

引用

International Conference on ASIC

作者： Shoudong Huang Wengao Lu Ye Zhou Shanzhe Yu Yacong Zhang Xueyou Shi Zhongjian Chen Key Laboratory of Microelectronic Devices and Circuits Institute of Microelectronics Peking University Beijing China

This paper presents a ramp generator with automatic slope calibration which is able to compensate slope deviation due to process and environmental effects. A novel analog feedback circuit is proposed to regulate the stop voltage in successive approximation while fix the start voltage of the ramp automatically without much calibration code input or complicated auxiliary circuits, which reduces the complexity of circuitry and timing, meanwhile, saves area and power. A 12-bit single-slope ADC (SS-ADC), employing the proposed ramp generator, has been implemented in a 0.11-μm CMOS process. The simulation results show that the offset of the ramp's stop voltage is reduced to 0.1 LSB and the nonlinearity of the ramp is 0.034%. Besides, the measurement results verify that this structure can fulfill automatic slope calibration.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Erratum to: CMN: a co-designed neural architecture search for efficient computing-in-memory-based mixture-of-experts

引用

Science China Information Sciences 2024年第11期67卷 1-1页

作者： Han, Shihao Liu, Sishuo Du, Shucheng Li, Mingzi Ye, Zijian Xu, Xiaoxin Li, Yi Wang, Zhongrui Shang, Dashan Department of Electrical and Electronic Engineering The University of Hong Kong Hong Kong China ACCESS-AI Chip Center for Emerging Smart Systems InnoHK Centers Hong Kong Science Park Hong Kong China Key Lab of Fabrication Technologies for Integrated Circuits Chinese Academy of Sciences Beijing China Laboratory of Microelectronics Devices and Integrated Technology Institute of Microelectronics Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China School of Microelectronics Southern University of Science and Technology Shenzhen China

来源：评论

学校读者我要写书评

暂无评论

In-memory search with learning to hash based on resistive memory for recommendation acceleration

npj Unconventional Computing

引用

npj Unconventional Computing 2024年第1期1卷 1-9页

作者： Fei Wang Woyu Zhang Zhi Li Rui Bao Xiaoxin Xu Chunmeng Dou Dashan Shang Ning Lin Zhongrui Wang Key Lab of Fabrication Technologies for Integrated Circuits Institute of Microelectronics Chinese Academy of Sciences Beijing 100029 China Key Laboratory of Microelectronic Devices and Integrated Technology Institute of Microelectronics Chinese Academy of Sciences Beijing 100029 China University of Chinese Academy of Sciences Beijing 100049 China Department of Electrical and Engineering the University of Hong Kong Hong Kong China ACCESS - AI Chip Center for Emerging Smart Systems InnoHK Centers Hong Kong Science Park Hong Kong China School of Microelectronics Southern University of Science and Technology Shenzhen 518055 China

Similarity search is essential in current artificial intelligence applications and widely utilized in various fields, such as recommender systems. However, the exponential growth of data poses significant challenges in search time and energy consumption on traditional digital hardware. Here, we propose a software-hardware co-optimization to address these challenges. On the software side, we employ a learning-to-hash method for vector encoding and achieve an approximate nearest neighbor search by calculating Hamming distance, thereby reducing computational complexity. On the hardware side, we leverage the resistance random-access memory crossbar array to implement the hash encoding process and the content-addressable memory with an in-memory computing paradigm to lower the energy consumption during searches. Simulations on the MovieLens dataset demonstrate that the implementation achieves comparable accuracy to software and reduces energy consumption by 30-fold compared to traditional digital systems. These results provide insight into the development of energy-efficient in-memory search systems for edge computing.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Corrigendum to “Epitaxial growth of wafer scale antioxidant single-crystal graphene on twinnedPt(111)” [Carbon 181 (2021) 225-233]

引用

Carbon 2021年 182卷 863-864页

作者： He Kang Pengtao Tang Haibo Shu Yanhui Zhang Yijian Liang Jing Li Zhiying Chen Yanping Sui Shike Hu Shuang Wang Sunwen Zhao Xuefu Zhang Chengxin Jiang Yulong Chen Zhongying Xue Miao Zhang Da Jiang Guanghui Yu Songang Peng Zhi Jin Xinyu Liu State Key Laboratory of Functional Materials for Informatics Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Shanghai 200050 China Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of Sciences Beijing 100049 China College of Optical and Electronic Technology China Jiliang University 310018 Hangzhou China Microwave Devices and Integrated Circuits Department Institute of Microelectronics Chinese Academy of Sciences Beijing 100029 China

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：