NN-Thunder: HW/SW Codesign for Accelerating DNNs with Heterogeneous Beyond-von Neumann Architectures
NN-Thunder:利用异构超越冯·诺依曼架构加速 DNN 的硬件/软件协同设计
基本信息
- 批准号:506419033
- 负责人:
- 金额:--
- 依托单位:
- 依托单位国家:德国
- 项目类别:Research Grants
- 财政年份:
- 资助国家:德国
- 起止时间:
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
Deep learning and deep neural networks (DNNs) have been adopted in many application areas and have a great potential to reshape the future of humankind. However, recent studies have demonstrated that classical von Neumann architectures are inherently inefficient when it comes to deep learning because such heavy workloads result in an extraordinary shift towards data-centric computing. Under such scenarios, a significant amount of energy is inevitably spent in moving a massive amount of data back and forth between processing elements and memory blocks. In this project, we envision a heterogeneous underlying HW, that combines various types of accelerators from both von Neumann and beyond-von Neumann architectures, offering a wide range of tradeoffs between computation accuracy, consumed energy, latency, and area footprint. Our project intends to address the limitation and improvement of beyond-von Neumann architectures w.r.t. performance and energy efficiency. The lack of fundamental exploration of suitable neural networks on beyond-von Neumann architectures hinders the possibility to exploit the full potential of such accelerators. In this regard, binarized neural networks (BNNs) offer the possibility of ultra-efficient hardware implementation and outstanding synergy with novel beyond-von Neumann architectures that are built from emerging beyond-CMOS devices. This project proposal plans to explore different means w.r.t. modeling, design, and optimization, to fully unleash the power of beyond-von Neumann neural network accelerators with the following goals: (1) cross-layer modeling for emerging technologies and abstractions of LiM (Logic-in-Memory) and PiM (Processing-in-Memory) for BNNs, (2) hardware-aware NN optimization for beyond-von Neumann architecture, and (3) HW/SW Codesign and Optimization. All in all, realizing a heterogeneous HW architecture is a key for future deep learning. It enables an ultra-efficient execution of hybrid-precision neural networks through HW/SW codesign and effectively allows the possibility to explore different novel tradeoffs.
深度学习和深度神经网络(DNN)已被应用于许多应用领域,并具有重塑人类未来的巨大潜力。然而,最近的研究表明,在深度学习方面,经典的冯·诺依曼架构本质上是低效的,因为如此繁重的工作负载导致了向以数据为中心的计算的巨大转变。在这种情况下,大量的能量不可避免地花费在处理元件和存储器块之间来回移动大量数据上。在这个项目中,我们设想了一个异构的底层硬件,它结合了来自冯·诺依曼和超越冯·诺依曼架构的各种类型的加速器,在计算精度、能耗、延迟和占地面积之间提供了广泛的权衡。我们的项目旨在解决的限制和改进的超越冯诺依曼架构w.r.t.性能和能效。缺乏对超越冯·诺依曼架构的合适神经网络的基本探索,阻碍了利用这种加速器的全部潜力的可能性。在这方面,二值化神经网络(BNN)提供了超高效硬件实现的可能性,以及与新兴超CMOS器件构建的新型超冯·诺依曼架构的出色协同作用。本项目建议书计划探索不同的方法,建模,设计和优化,以充分释放超越冯诺依曼神经网络加速器的力量,具有以下目标:(1)新兴技术的跨层建模和BNN的LiM(内存逻辑)和PiM(内存处理)的抽象,(2)超越冯诺依曼架构的硬件感知NN优化,以及(3)硬件/软件协同设计和优化。总而言之,实现异构硬件架构是未来深度学习的关键。它通过硬件/软件协同设计实现了混合精度神经网络的超高效执行,并有效地允许探索不同的新权衡。
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 }}
Professor Dr.-Ing. Hussam Amrouch, Ph.D.其他文献
Professor Dr.-Ing. Hussam Amrouch, Ph.D.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('Professor Dr.-Ing. Hussam Amrouch, Ph.D.', 18)}}的其他基金
ACCROSS: Approximate Computing aCROs the System Stack
ACCROSS:系统堆栈的近似计算 aCRO
- 批准号:
428566201 - 财政年份:
- 资助金额:
-- - 项目类别:
Research Grants
相似国自然基金
基于新一代致动器THUNDER的复杂封闭空间噪声多目标控制机理研究
- 批准号:50375027
- 批准年份:2003
- 资助金额:24.0 万元
- 项目类别:面上项目
相似海外基金
‘Like the Thunder’: Seeing Stories in the Gulf Country
“如雷霆”:在海湾国家看故事
- 批准号:
IN240100137 - 财政年份:2024
- 资助金额:
-- - 项目类别:
Discovery Indigenous
THUNDER - THermochemical storage Utilization eNabling Data centre seasonal Energy Recovery
THUNDER - 热化学存储利用 eNabling 数据中心季节性能源回收
- 批准号:
10088548 - 财政年份:2024
- 资助金额:
-- - 项目类别:
EU-Funded
Thunder Bay's Joint Mobile Crisis Response Team: Evaluation of Process, Outcomes and Costs Quality of Life and Care Experience among Complex Cancer Patients
Thunder Bay 联合流动危机应对小组:评估复杂癌症患者的过程、结果和成本生活质量和护理体验
- 批准号:
418307 - 财政年份:2020
- 资助金额:
-- - 项目类别:
Miscellaneous Programs
The Structure of the World as Viewed from Sanskrit Etymology: The Forms of the Fire, Thunder, and Sun Gods
从梵语词源看世界的结构:火神、雷神和太阳神的形态
- 批准号:
19K23012 - 财政年份:2019
- 资助金额:
-- - 项目类别:
Grant-in-Aid for Research Activity Start-up
"From Titanic to Tomorrow: Unlocking Genetic Secrets of the Past, Present and Future", A Café Scientifique series in Sudbury and Thunder Bay, Ontario
“从泰坦尼克号到明天:解锁过去、现在和未来的基因秘密”,安大略省萨德伯里和桑德贝的科学咖啡馆系列
- 批准号:
236962 - 财政年份:2011
- 资助金额:
-- - 项目类别:
Miscellaneous Programs
International Joint Observation of Thunder stroms in Iinter Tropical Convergence Zone
热带辐合带雷暴国际联合观测
- 批准号:
10041142 - 财政年份:1998
- 资助金额:
-- - 项目类别:
Grant-in-Aid for Scientific Research (B).
Characteristic of Lightning Channel in the Thundercloud from Thunder Measurement
雷电测量中雷云中闪电通道的特征
- 批准号:
05650287 - 财政年份:1993
- 资助金额:
-- - 项目类别:
Grant-in-Aid for General Scientific Research (C)
Characteristic of Lightning Channel in the Thundercloud from Thunder
雷电云中闪电通道的特征
- 批准号:
02650210 - 财政年份:1990
- 资助金额:
-- - 项目类别:
Grant-in-Aid for General Scientific Research (C)
Numerical Simulation and Observational Analysis of Thunder- storms and Subcloud Phenomena
雷暴和亚云现象的数值模拟与观测分析
- 批准号:
8604402 - 财政年份:1986
- 资助金额:
-- - 项目类别:
Continuing Grant