CAREER: Scaling-up Resistive Synaptic Arrays for Neuro-inspired Computing

职业:扩大电阻突触阵列以实现神经启发计算

基本信息

  • 批准号:
    1903951
  • 负责人:
  • 金额:
    $ 29.38万
  • 依托单位:
  • 依托单位国家:
    美国
  • 项目类别:
    Continuing Grant
  • 财政年份:
    2018
  • 资助国家:
    美国
  • 起止时间:
    2018-07-01 至 2022-08-31
  • 项目状态:
    已结题

项目摘要

Neuro-inspired deep learning algorithms have demonstrated their power in executing intelligent tasks such as image and speech recognition. However, training of such deep neural networks requires huge amount of computational resources that are not affordable for mobile applications. Hardware acceleration of deep learning, with orders of magnitude improvement in speed and energy efficiency, remains a grand challenge for the conventional hardware based on silicon CMOS technology and von-Neumann architecture. As the learning algorithms extensively involve matrix operations, neuro-inspired architectures that leverage the distributed computing in the neuron nodes and localized storage in the synaptic networks are very attractive. The ultimate goal of this project is to advance the neuro-inspired computing with emerging nano-device technologies towards a self-learning chip. A chip that learns in real-time and consumes low-power can be placed at frontend sensors, bringing broad benefits for a number of current applications. The PI will establish close collaboration with industry through student internships and technology transfer. The plan for integration of research and education will train students with interdisciplinary skills. The cross-layer nature of this project ranging from semiconductor device, circuit design, electronic design automation, and machine learning is expected to provide an ideal platform for this educational goal.The technical goal of this project is to overcome the challenges that prevent scaling up of the crossbar array size for neuro-inspired architecture. Resistive devices with continuous multilevel states have been proposed to function as synaptic weights in the crossbar architecture. However, with the increase of the array size, issues associated with device yield, device variability, and array parasitics will arise and may degrade the system performance. The PI plans to tackle these challenges by exploiting hierarchical research efforts from devices, circuits and architectures. The outcome of the research includes device compact model, circuit-level benchmark simulator for estimating the area/latency/power of the crossbar array macro, and architectural tool for efficiently mapping the learning algorithms into the crossbar architecture. The PI has established a custom fabrication channel for tape-out of resistive devices on top of CMOS peripheral circuits via his collaboration with academic partners. The prototype chip with measured data is expected to make a strong impact on this field, which previously relied on the simulations for predicting large-scale array performance.
神经启发的深度学习算法已经证明了它们在执行图像和语音识别等智能任务方面的能力。然而,这种深度神经网络的训练需要大量的计算资源,这对于移动应用来说是无法承受的。深度学习的硬件加速在速度和能源效率上都有数量级的提高,这对于基于硅CMOS技术和冯-诺伊曼架构的传统硬件来说仍然是一个巨大的挑战。由于学习算法广泛涉及矩阵运算,利用神经元节点的分布式计算和突触网络的局部存储的神经启发架构非常有吸引力。该项目的最终目标是利用新兴的纳米设备技术将神经启发计算推向自我学习芯片。一个实时学习和低功耗的芯片可以放置在前端传感器上,为当前的许多应用带来广泛的好处。PI将通过学生实习和技术转让与工业界建立密切合作关系。研究与教育相结合的计划将培养学生的跨学科技能。这个项目的跨层性质涵盖了半导体器件、电路设计、电子设计自动化和机器学习,有望为这一教育目标提供理想的平台。该项目的技术目标是克服阻碍神经启发建筑的交叉杆阵列尺寸扩大的挑战。具有连续多能级状态的电阻器件已被提出作为交叉杆结构中的突触权重。然而,随着阵列大小的增加,与器件良率、器件可变性和阵列寄生相关的问题将会出现,并可能降低系统性能。PI计划通过利用设备、电路和架构的分层研究成果来应对这些挑战。研究成果包括器件紧凑模型,用于估计交叉棒阵列宏的面积/延迟/功率的电路级基准模拟器,以及用于有效地将学习算法映射到交叉棒架构中的架构工具。通过与学术合作伙伴的合作,PI已经建立了在CMOS外围电路之上的带出电阻器件的定制制造渠道。具有测量数据的原型芯片有望对该领域产生重大影响,此前该领域依赖于预测大规模阵列性能的模拟。

项目成果

期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ monograph.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ sciAawards.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ conferencePapers.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ patent.updateTime }}

Shimeng Yu其他文献

Optimization of RRAM-Based Physical Unclonable Function With a Novel Differential Read-Out Method
采用新颖的差分读出方法优化基于 RRAM 的物理不可克隆功能
  • DOI:
    10.1109/led.2016.2647230
  • 发表时间:
    2017-02
  • 期刊:
  • 影响因子:
    4.9
  • 作者:
    Yachuan Pang;Huaqiang Wu;Bin Gao;Ning Deng;Dong Wu;Rui Liu;Shimeng Yu;An Chen;He Qian
  • 通讯作者:
    He Qian
First Experimental Demonstration of Robust HZO/β-Ga₂O₃ Ferroelectric Field-Effect Transistors as Synaptic Devices for Artificial Intelligence Applications in a High-Temperature Environment
鲁棒 HZO/β-Ga2O3 铁电场效应晶体管作为高温环境下人工智能应用突触器件的首次实验演示
  • DOI:
  • 发表时间:
    2021
  • 期刊:
  • 影响因子:
    3.1
  • 作者:
    J. Noh;H. Bae;Junkang Li;Yandong Luo;Y. Qu;T. J. Park;M. Si;Xuegang Chen;A. Charnas;W. Chung;Xiaochen Peng;S. Ramanathan;Shimeng Yu;P. Ye
  • 通讯作者:
    P. Ye
Resistive Random Access Memory (RRAM)
  • DOI:
    10.1007/978-3-031-02030-8
  • 发表时间:
    2016
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Shimeng Yu
  • 通讯作者:
    Shimeng Yu
Ferroelectric FET based Non-Volatile Analog Synaptic Weight Cell
基于铁电 FET 的非易失性模拟突触重量单元
  • DOI:
  • 发表时间:
    2019
  • 期刊:
  • 影响因子:
    0
  • 作者:
    M. Jerry;S. Dutta;K. Ni;Jianchi Zhang;Pankaj Sharma;S. Datta;A. Kazemi;X. Hu;M. Niemier;Pai;Shimeng Yu
  • 通讯作者:
    Shimeng Yu
A phenomenological model of oxygen ion transport for metal oxide resistive switching memory
金属氧化物阻变存储器氧离子传输的唯象模型

Shimeng Yu的其他文献

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

{{ truncateString('Shimeng Yu', 18)}}的其他基金

Collaborative Research: FET: Medium:Compact and Energy-Efficient Compute-in-Memory Accelerator for Deep Learning Leveraging Ferroelectric Vertical NAND Memory
合作研究:FET:中型:紧凑且节能的内存计算加速器,用于利用铁电垂直 NAND 内存进行深度学习
  • 批准号:
    2312885
  • 财政年份:
    2023
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
Low Temperature Embedded Memory Devices for Near-Memory and In-Memory Computing
用于近内存和内存计算的低温嵌入式存储器件
  • 批准号:
    2218604
  • 财政年份:
    2022
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
Exploiting Metal-Insulator-Transition in Strongly Correlated Oxides as Neuron Device for Neuro-Inspired Computing
利用强相关氧化物中的金属-绝缘体转变作为神经元设备进行神经启发计算
  • 批准号:
    1903577
  • 财政年份:
    2018
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
STARSS: Small: Design of Light-weight RRAM based Hardware Security Primitives for IoT devices
STARSS:小型:为物联网设备设计基于 RRAM 的轻量级硬件安全原语
  • 批准号:
    1903631
  • 财政年份:
    2018
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
Exploiting Metal-Insulator-Transition in Strongly Correlated Oxides as Neuron Device for Neuro-Inspired Computing
利用强相关氧化物中的金属-绝缘体转变作为神经元设备进行神经启发计算
  • 批准号:
    1701565
  • 财政年份:
    2017
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
CAREER: Scaling-up Resistive Synaptic Arrays for Neuro-inspired Computing
职业:扩大电阻突触阵列以实现神经启发计算
  • 批准号:
    1552687
  • 财政年份:
    2016
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
STARSS: Small: Design of Light-weight RRAM based Hardware Security Primitives for IoT devices
STARSS:小型:为物联网设备设计基于 RRAM 的轻量级硬件安全原语
  • 批准号:
    1615774
  • 财政年份:
    2016
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
EAGER: Monolithic 3D Integration of Resistive Random Access Memory (ReRAM): A Technological Exploration
EAGER:电阻式随机存取存储器 (ReRAM) 的单片 3D 集成:技术探索
  • 批准号:
    1449653
  • 财政年份:
    2014
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant

相似海外基金

CAREER: Scaling up Brain Circuit Reconstruction with Human-centric Machine Learning
职业:利用以人为本的机器学习扩大脑回路重建
  • 批准号:
    2239688
  • 财政年份:
    2023
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling-up effects of biodiversity on ecosystem productivity: from local communities to landscapes
职业:扩大生物多样性对生态系统生产力的影响:从当地社区到景观
  • 批准号:
    1845334
  • 财政年份:
    2019
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling Up Knowledge Discovery in High-Dimensional Data Via Nonconvex Statistical Optimization
职业:通过非凸统计优化扩大高维数据中的知识发现
  • 批准号:
    1906169
  • 财政年份:
    2018
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling Up Mobile Accessibility Through Touchscreen Personalization
职业:通过触摸屏个性化扩展移动辅助功能
  • 批准号:
    1818594
  • 财政年份:
    2017
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling Up Knowledge Discovery in High-Dimensional Data Via Nonconvex Statistical Optimization
职业:通过非凸统计优化扩大高维数据中的知识发现
  • 批准号:
    1652539
  • 财政年份:
    2017
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling-up Resistive Synaptic Arrays for Neuro-inspired Computing
职业:扩大电阻突触阵列以实现神经启发计算
  • 批准号:
    1552687
  • 财政年份:
    2016
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling Up Mobile Accessibility Through Touchscreen Personalization
职业:通过触摸屏个性化扩展移动辅助功能
  • 批准号:
    1350438
  • 财政年份:
    2014
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
CAREER: Scaling up Modeling and Statistical Inference for Massive Collections of Time Series
职业:扩大大规模时间序列集合的建模和统计推断
  • 批准号:
    1350133
  • 财政年份:
    2014
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
Forming Better STEM Career Trajectories: Sustaining and Scaling-Up CAP
形成更好的 STEM 职业轨迹:维持和扩大 CAP
  • 批准号:
    1316702
  • 财政年份:
    2013
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Standard Grant
CAREER: Scaling Up First-Order Logical Reasoning with Graphical Structure
职业:用图形结构扩展一阶逻辑推理
  • 批准号:
    0546663
  • 财政年份:
    2005
  • 资助金额:
    $ 29.38万
  • 项目类别:
    Continuing Grant
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了