.png){kind=icon}
返回
Approximate Down-Sampling Strategy for Power-Constrained Intelligent Systems
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2022.3142292.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In modern power constrained applications, as with most of those belonging to the Internet-of-Things world, custom hardware supports are ever more commonly adopted to deploy artificial intelligence algorithms. In these operating environments, limiting the power dissipation as much as possible is mandatory, even at the expense of reduced computational accuracy. In this paper we propose a novel prediction method to identify potential predominant features in convolutional layers followed by down-sampling layers, thus reducing the overall number of convolution calculations. This approximation down-sampling strategy has been exploited to design a custom hardware architecture for the inference of Convolutional Neural Network (CNN) models. The proposed approach has been applied to several benchmark CNN models and we achieved an overall energy saving of up to 70% with an accuracy loss lower than 3%, with respect to baseline designs. Performed experiments demonstrate that, when adopted to infer the Visual Geometry Group-16 (VGG16) network model, the proposed architecture implemented on a Xilinx Z-7045 chip and on the STM 28nm process technology dissipates only 680 and 21.9 mJ/frame, respectively. In both cases, the novel design overcomes several state-of-the-art competitors in terms of energy-accuracy drop product.
Keyword:
Computer architecture
Quantization (signal)
Convolution
Hardware
Filtering algorithms
Convolutional neural networks
Prediction algorithms
Approximate computing
convolutional neural networks
low-power hardware architectures
pooling layers
期刊
IF:
3.6
论文数:
9.8W
被引数:
29.4W
机构
引用论文
SDK Reinvented: Document Image Analysis Methods as RESTful Web ServicesSDK重塑：文档图像分析方法作为RESTful Web服务
PREAI
分享收藏An Energy-Efficient Implementation of Group Pruned CNNs on FPGA群剪枝cnn在FPGA上的节能实现.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏Design and Development of a Deep Learning-Based Model for Anomaly Detection in IoT Networks基于深度学习的物联网网络异常检测模型的设计与开发IEEE ACCESSOAAI
IEEE ACCESSIF3.6
OAAI分享收藏Learning hand-eye coordination for robotic grasping with deep learning and large-scale data collection通过深度学习和大规模数据收集学习机器人抓取的手眼协调OAAI
OAAI分享收藏Structural study of lanthanides(III) in aqueous nitrate and chloride solutions by EXAFS通过EXAFS对硝酸盐和氯化物水溶液中镧系元素 (III) 的结构研究
PREAI
分享收藏
查看更多内容