BriefGPT.xyz
Apr, 2018
卷积神经网络的加速器感知剪枝
Accelerator-Aware Pruning for Convolutional Neural Networks
HTML
PDF
Hyeong-Ju Kang
TL;DR
在嵌入式环境中,卷积神经网络因其过多的权重存储和算术运算而未能得到广泛应用,为解决这一问题,本文提出了一种新的修剪方案,以反映加速器架构,通过此方案,性能得到了大幅提升,并成功应用于AlexNet,VGG16,ResNet,MobileNet等多种网络模型。
Abstract
convolutional neural networks
have shown tremendous performance in computer vision tasks,but their excessive amount of weights and operations prevent them from being adopted in embedded environments. One of the solutions involves
→