CAREER: Efficient Algorithms and Hardware for Accelerated Machine Learning

职业:用于加速机器学习的高效算法和硬件

基本信息

  • 批准号:
    1943349
  • 负责人:
  • 金额:
    $ 50万
  • 依托单位:
  • 依托单位国家:
    美国
  • 项目类别:
    Continuing Grant
  • 财政年份:
    2020
  • 资助国家:
    美国
  • 起止时间:
    2020-06-15 至 2025-05-31
  • 项目状态:
    未结题

项目摘要

Artificial Intelligence (AI) is bringing significant benefits to human society with outstanding potential to address essential human concerns in the future. Deep neural networks have shown significant improvements in many AI applications. However, the benefit comes at the high cost of computational resources and engineer resources. For mobile devices with a tight power budget, such high demands for computation become prohibitive. On the other side, there is a shortage of experts who can design neural networks, which are notoriously hard to tune. This project systematically investigates efficient neural network architectures and their hardware accelerators to make them run fast at low power. It also aims to accelerate the design cycle by AI-based design automation (AI-designed AI). Such design methodology can support machine learning research and education,, while significantly improving the productivity of machine learning models. People no longer have to hand-tune the model, but it is automated, enabling non-experts to build efficient machine learning models. It will democratize AI to a more diverse community. The project aims to auto-generate both efficient neural networks and their hardware implementations that can generalize to high-dimensional representations, through automatic machine learning (AutoML) techniques. It will design a hardware accelerator to provide more computation per unit cost. The algorithm hardware co-design approach unveils a larger design space that conventional wisdom has been limited to. It is expected to shorten the design cycle of neural architecture search by two orders of magnitude over existing work. With AutoML, powerful hardware, and the co-designed efficient algorithms, it is possible to solve more challenging AI tasks on high-dimensional data that are previously difficult or impossible, hindered by the computation resource, such as videos and 3D point clouds. These techniques give rise to a deeper understanding of the high dimensional representations and produce state-of-the-art neural network architectures that can efficiently run on mobile devices as well as protect user privacy.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
人工智能(AI)正在为人类社会带来巨大的利益,在解决未来人类关注的基本问题方面具有突出的潜力。深度神经网络在许多人工智能应用中都显示出了显著的改进。然而,好处是以计算资源和工程资源的高成本为代价的。对于电力预算紧张的移动设备来说,如此高的计算要求变得令人望而却步。另一方面,缺乏能够设计神经网络的专家,神经网络的调整是出了名的难。本项目系统地研究了高效的神经网络体系结构及其硬件加速器,以使其在低功耗下快速运行。它还旨在通过基于人工智能的设计自动化(AI-Designed AI)来加快设计周期。这种设计方法可以支持机器学习研究和教育,同时显著提高机器学习模型的生产率。人们不再需要手动调整模型,但它是自动的,使非专家能够构建高效的机器学习模型。它将使人工智能民主化,形成一个更加多样化的社区。该项目旨在通过自动机器学习(AutoML)技术,自动生成高效的神经网络及其硬件实现,这些硬件实现可以推广到高维表示。它将设计一个硬件加速器,以提供更多的单位成本计算。算法硬件协同设计方法揭示了一个更大的设计空间,而传统智慧一直局限于此。它有望将神经结构搜索的设计周期比现有工作缩短两个数量级。有了AutoML、强大的硬件和共同设计的高效算法,就有可能解决高维数据上更具挑战性的人工智能任务,这些任务以前是困难或不可能的,受到计算资源的阻碍,如视频和3D点云。这些技术带来了对高维表示的更深层次的理解,并产生了可以在移动设备上高效运行并保护用户隐私的最先进的神经网络结构。该奖项反映了NSF的法定使命,并通过使用基金会的智力优势和更广泛的影响审查标准进行评估,被认为值得支持。

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 }}

Song Han其他文献

Preparation, Characterization of Phosphorus Doped Titania Photocatalysts with High Photocatalystic Properties
高光催化性能磷掺杂二氧化钛光催化剂的制备及表征
  • DOI:
    10.4028/www.scientific.net/amr.113-116.2154
  • 发表时间:
    2010
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Siyao Guo;J. Sun;F. Wang;Lin Yang;Feng Zhang;Song Han
  • 通讯作者:
    Song Han
Expansion strain model and damage risk control for cement-based materials with low water–binder ratios under rehydration
低水胶比水泥基材料复水膨胀应变模型及损伤风险控制
  • DOI:
    10.1016/j.conbuildmat.2021.122996
  • 发表时间:
    2021-06
  • 期刊:
  • 影响因子:
    7.4
  • 作者:
    Yazhou Liu;Mingzhe An;Ge Zhang;Ziruo Yu;Yue Wang;Song Han
  • 通讯作者:
    Song Han
Study on NOxEmission Reduction in Coke Combustion and Sintering Process
焦炭燃烧及烧结过程NOx减排研究
  • DOI:
    10.3103/s1068364x19120093
  • 发表时间:
    2019
  • 期刊:
  • 影响因子:
    0.4
  • 作者:
    Song Han;Lin Dong;Zhiping Lei;Aiming Ke;Con Shi;Jing Chong Yan;Zhanku Li;Shigang Kang;Hengfu Shui;Zhicai Wang;Shibiao Ren;Chunxiu Pan
  • 通讯作者:
    Chunxiu Pan
Improved predictive functional control for ethylene cracking furnace
乙烯裂解炉改进的预测功能控制
  • DOI:
    10.1177/0020294019842602
  • 发表时间:
    2019-04
  • 期刊:
  • 影响因子:
    2
  • 作者:
    Song Han;Su Cheng-li;Shi Hui-yuan;Li Ping;Cao Jiang-tao
  • 通讯作者:
    Cao Jiang-tao
Hydroisomerization of n-hexane over gallium-promoted sulfated zirconia
镓促进的硫酸化氧化锆上正己烷的加氢异构化
  • DOI:
    10.1016/j.catcom.2003.08.003
  • 发表时间:
    2003
  • 期刊:
  • 影响因子:
    3.7
  • 作者:
    C. Cao;Song Han;Changlin Chen;N. Xu;Chunye Mou
  • 通讯作者:
    Chunye Mou

Song Han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('Song Han', 18)}}的其他基金

Collaborative Research: SHF: Medium: Heterogeneous Architecture for Collaborative Machine Learning
协作研究:SHF:媒介:协作机器学习的异构架构
  • 批准号:
    2106711
  • 财政年份:
    2021
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
Collaborative Research: PPoSS: LARGE: Principles and Infrastructure of Extreme Scale Edge Learning for Computational Screening and Surveillance for Health Care
合作研究:PPoSS:大型:用于医疗保健计算筛查和监视的超大规模边缘学习的原理和基础设施
  • 批准号:
    2119340
  • 财政年份:
    2021
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
Collaborative Research: PPoSS: Planning: S3-IoT: Design and Deployment of Scalable, Secure, and Smart Mission-Critical IoT Systems
协作研究:PPoSS:规划:S3-IoT:可扩展、安全和智能的关键任务物联网系统的设计和部署
  • 批准号:
    2028875
  • 财政年份:
    2020
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant
Collaborative Research: PPoSS: Planning: Principles for Edge Sensing and Computing for Personalized, Precision Healthcare at National Scale
合作研究:PPoSS:规划:全国范围内个性化精准医疗的边缘传感和计算原则
  • 批准号:
    2028888
  • 财政年份:
    2020
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant
RAPID: Preventing the Spread of Coronavirus with Efficient Deep Learning
RAPID:通过高效的深度学习防止冠状病毒的传播
  • 批准号:
    2027266
  • 财政年份:
    2020
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant
CNS Core: Small: Dynamic and Composite Resource Management in Large-scale Industrial IoT Systems
CNS 核心:小型:大型工业物联网系统中的动态复合资源管理
  • 批准号:
    2008463
  • 财政年份:
    2020
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant
CPS: Small: Collaborative Research: A Secure Communication Framework with Verifiable Authenticity for Immutable Services in Industrial IoT Systems
CPS:小型:协作研究:工业物联网系统中不可变服务的具有可验证真实性的安全通信框架
  • 批准号:
    1932480
  • 财政年份:
    2019
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant
PFI-TT: Developing a Configurable Real-time High-speed Wireless Communication Platform for Large-scale Industrial Control Systems
PFI-TT:为大型工业控制系统开发可配置的实时高速无线通信平台
  • 批准号:
    1919229
  • 财政年份:
    2019
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant
CCRI: Planning: Collaborative Research: A Software-defined Wireless Communications Network Research Infrastructure for the Industrial Internet of Things(IIoT)Research Community
CCRI:规划:协作研究:工业物联网(IIoT)研究社区的软件定义无线通信网络研究基础设施
  • 批准号:
    1925706
  • 财政年份:
    2019
  • 资助金额:
    $ 50万
  • 项目类别:
    Standard Grant

相似海外基金

CAREER: Blessing of Nonconvexity in Machine Learning - Landscape Analysis and Efficient Algorithms
职业:机器学习中非凸性的祝福 - 景观分析和高效算法
  • 批准号:
    2337776
  • 财政年份:
    2024
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Efficient Algorithms for Modern Computer Architecture
职业:现代计算机架构的高效算法
  • 批准号:
    2339310
  • 财政年份:
    2024
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: A Theoretical Exploration of Efficient and Accurate Clustering Algorithms
职业生涯:高效准确聚类算法的理论探索
  • 批准号:
    2337832
  • 财政年份:
    2024
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Computation-efficient Algorithms for Grid-scale Energy Storage Control, Bidding, and Integration Analysis
职业:用于电网规模储能控制、竞价和集成分析的计算高效算法
  • 批准号:
    2239046
  • 财政年份:
    2023
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Efficient Uncertainty Quantification in Turbulent Combustion Simulations: Theory, Algorithms, and Computations
职业:湍流燃烧模拟中的高效不确定性量化:理论、算法和计算
  • 批准号:
    2143625
  • 财政年份:
    2022
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: From Shallow to Deep Representation Learning: Global Nonconvex Optimization Theories and Efficient Algorithms
职业:从浅层到深层表示学习:全局非凸优化理论和高效算法
  • 批准号:
    2143904
  • 财政年份:
    2022
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Genomic Data Science: From Informational Limits to Efficient Algorithms
职业:基因组数据科学:从信息限制到高效算法
  • 批准号:
    2046991
  • 财政年份:
    2021
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Robust and Efficient Algorithms for Statistical Estimation and Inference
职业:用于统计估计和推理的稳健且高效的算法
  • 批准号:
    2045068
  • 财政年份:
    2021
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Efficient Fine-grained Algorithms
职业:高效的细粒度算法
  • 批准号:
    2223282
  • 财政年份:
    2021
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
CAREER: Efficient and Accurate Local Time-Stepping Algorithms for Multiscale Multiphysics Systems
职业:多尺度多物理系统的高效、准确的局部时间步进算法
  • 批准号:
    2041884
  • 财政年份:
    2021
  • 资助金额:
    $ 50万
  • 项目类别:
    Continuing Grant
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了