首页> 美国卫生研究院文献>Sensors (Basel Switzerland) >Mapping Neural Networks to FPGA-Based IoT Devices for Ultra-Low Latency Processing
【2h】

Mapping Neural Networks to FPGA-Based IoT Devices for Ultra-Low Latency Processing

机译:将神经网络映射到基于FPGA的IoT设备以进行超低延迟处理

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务,本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文,但由于OA文献来源多样且变更频繁,仍可能出现获取不到、文献不完整或与标题不符等情况,如果获取不到我们将提供退款服务。请知悉。

摘要

Internet of things (IoT) infrastructure, fast access to knowledge becomes critical. In some application domains, such as robotics, autonomous driving, predictive maintenance, and anomaly detection, the response time of the system is more critical to ensure Quality of Service than the quality of the answer. In this paper, we propose a methodology, a set of predefined steps to be taken in order to map the models to hardware, especially field programmable gate arrays (FPGAs), with the main focus on latency reduction. Multi-objective covariance matrix adaptation evolution strategy (MO-CMA-ES) was employed along with custom scores for sparsity, bit-width of the representation and quality of the model. Furthermore, we created a framework which enables mapping of neural models to FPGAs. The proposed solution is validated using three case studies and Xilinx Zynq UltraScale+ MPSoC 285 XCZU15EG as a platform. The results show a compression ratio for quantization and pruning in different scenarios with and without retraining procedures. Using our publicly available framework, we achieved 210 ns of latency for a single processing step for a model composed of two long short-term memory (LSTM) and a single dense layer.
机译:物联网(IoT)基础架构,快速获取知识变得至关重要。在某些应用领域中,例如机器人技术,自动驾驶,预测性维护和异常检测,对于确保服务质量而言,系统的响应时间比答案的质量更为关键。在本文中,我们提出了一种方法论,即要采取的一系列预定义步骤,以将模型映射到硬件,尤其是现场可编程门阵列(FPGA),主要侧重于减少延迟。采用多目标协方差矩阵适应进化策略(MO-CMA-ES)以及稀疏性,表示的位宽和模型质量的自定义分数。此外,我们创建了一个框架,该框架支持将神经模型映射到FPGA。所提出的解决方案已通过三个案例研究和Xilinx Zynq UltraScale + MPSoC 285 XCZU15EG作为平台进行了验证。结果显示了在有和没有再训练程序的情况下,在不同情况下进行量化和修剪的压缩率。使用我们的公共框架,对于由两个长短期内存(LSTM)和一个密集层组成的模型,单个处理步骤的延迟达到210 ns。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
代理获取

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号