Neural Approximate Accelerator Architecture Optimization for DNN Inference on Lightweight FPGAs (NA^3Os)

用于轻量级 FPGA (NA^3Os) 上 DNN 推理的神经近似加速器架构优化

基本信息

项目摘要

Deep learning has changed the way and the quality complex technical problems have been solved. Many of the advances from mostly within the last decade already had profound impact on systems used in everyday life. Background is that technical systems are often to such an extent complex that it is infeasible to build sufficiently accurate models that may serve as a basis for classical optimization techniques. These are the scenarios where Deep Neural Networks(DNNs) shine. The drawback, however, are high computational demands to process DNNs. Besides high computing requirements, memory resources and energy, among others, are often also very high. This proposal presents an approach to successfully deploy DNNs in systems with very limited resources, particularly FPGAs, enabling thus efficient TinyML implementation. An emphasis in our investigations is put on a unique amalgam (combination) of compression techniques such as pruning and quantization with emerging approximate computing principles. Particularly for FPGAs, we want to investigate the opportunity of approximate arithmetic units. Moreover, we want to exploit FPGA-specific artifacts such as DSPs and BRAMs to provide highly resource and energy efficient hardware implementations for DNNs. To the best of our knowledge, this proposal presents the first important steps in optimizing the deployment of DNNs on approximate and reconfigurable hardware. This involves investigating innovative mapping and design space exploration techniques. The combination of micro-architectural peculiarities with the approximate computing paradigm promises a controllable trade-off between the quality of DNN results and the computational resources needed. The final goal is the development of a co-search methodology between the neural network architecture, its optimization, and the synthesis of approximated DNN accelerators on FPGAs. Further research includes the analysis of DNN robustness and energy tradeoffs. In summary, we propose the first steps to successfully deploy DNN on highly resource-constrained FPGA systems while exploiting approximate computing principles.
深度学习改变了方式,解决了质量复杂的技术问题。过去十年的许多进步已经对日常生活中使用的系统产生了深远的影响。背景是,技术系统通常复杂到这样的程度,以至于不可能构建可以用作经典优化技术的基础的足够精确的模型。这些都是深度神经网络(DNN)大放异彩的场景。然而,缺点是处理DNN的高计算需求。除了高计算要求之外,存储器资源和能量等通常也非常高。该提案提出了一种在资源非常有限的系统中成功部署DNN的方法,特别是FPGA,从而实现高效的TinyML实现。在我们的调查重点放在一个独特的汞合金(组合)的压缩技术,如修剪和量化与新兴的近似计算原理。特别是对于FPGA,我们希望研究近似算术单元的机会。此外,我们希望利用FPGA特定的工件,如DSP和BRAM,为DNN提供高度资源和能源效率的硬件实现。据我们所知,该提案提出了在近似和可重构硬件上优化DNN部署的第一个重要步骤。这涉及研究创新的测绘和设计空间探索技术。微架构特性与近似计算范式的结合有望在DNN结果的质量和所需的计算资源之间实现可控的权衡。最终目标是在神经网络架构、其优化和FPGA上近似DNN加速器的合成之间开发一种共同搜索方法。进一步的研究包括DNN鲁棒性和能量权衡的分析。总之,我们提出了在高度资源受限的FPGA系统上成功部署DNN的第一步,同时利用近似计算原理。

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. Jörg Henkel其他文献

Professor Dr.-Ing. Jörg Henkel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. Jörg Henkel', 18)}}的其他基金

Coordination Funds
协调基金
  • 批准号:
    264242055
  • 财政年份:
    2014
  • 资助金额:
    --
  • 项目类别:
    Priority Programmes
VirTherm-3D - Formalization of Multi-Agent System Management and Adaptive Modular Redundancy for Dependable 3D MPSoCs
VirTherm-3D - 可靠 3D MPSoC 的多代理系统管理和自适应模块化冗余的形式化
  • 批准号:
    181739223
  • 财政年份:
    2010
  • 资助金额:
    --
  • 项目类别:
    Priority Programmes
Koordinatoren-Antrag SPP 1500
协调员申请 SPP 1500
  • 批准号:
    190788807
  • 财政年份:
    2010
  • 资助金额:
    --
  • 项目类别:
    Priority Programmes
OTERA-III: Online Test Strategies for Reliable Reconfigurable Architectures - From Reliability to Guaranteed System Performability: A Multi-Layer Approach
OTERA-III:可靠可重构架构的在线测试策略 - 从可靠性到保证系统性能:多层方法
  • 批准号:
    182065442
  • 财政年份:
    2010
  • 资助金额:
    --
  • 项目类别:
    Priority Programmes

相似海外基金

CAREER: Speedy and Reliable Approximate Queries in Hybrid Transactional/Analytical Systems
职业:混合事务/分析系统中快速可靠的近似查询
  • 批准号:
    2339596
  • 财政年份:
    2024
  • 资助金额:
    --
  • 项目类别:
    Continuing Grant
NAfANE: New Approaches for Approximate Nash Equilibria
NAfANE:近似纳什均衡的新方法
  • 批准号:
    EP/X039862/1
  • 财政年份:
    2024
  • 资助金额:
    --
  • 项目类别:
    Research Grant
Collaborative Research: OAC: Approximate Nearest Neighbor Similarity Search for Large Polygonal and Trajectory Datasets
合作研究:OAC:大型多边形和轨迹数据集的近似最近邻相似性搜索
  • 批准号:
    2313039
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Standard Grant
A study of SNN device using serial approximate adders
使用串行近似加法器的SNN装置的研究
  • 批准号:
    23K11034
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
Efficient simulation and inference under approximate models of ancestry
祖先近似模型下的高效模拟和推理
  • 批准号:
    EP/X022595/1
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Research Grant
Compilation and Verification of Quantum Software in the Noisy and Approximate Regime
嘈杂近似体系中量子软件的编译与验证
  • 批准号:
    EP/Y004736/1
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Research Grant
Approximate Truths: A New Ground for the Pillars of Scientific Realism
近似真理:科学实在论支柱的新基础
  • 批准号:
    2908312
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Studentship
Collaborative Research: CIF: Small: Approximate Coded Computing - Fundamental Limits of Precision, Fault-Tolerance, and Privacy
协作研究:CIF:小型:近似编码计算 - 精度、容错性和隐私的基本限制
  • 批准号:
    2231706
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Standard Grant
Collaborative Research: CIF: Small: Approximate Coded Computing - Fundamental Limits of Precision, Fault-tolerance and Privacy
协作研究:CIF:小型:近似编码计算 - 精度、容错性和隐私的基本限制
  • 批准号:
    2231707
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Standard Grant
Approximate Commutators and K-theory
近似换向器和 K 理论
  • 批准号:
    2247968
  • 财政年份:
    2023
  • 资助金额:
    --
  • 项目类别:
    Standard Grant
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了