卷积神经网络的硬件逼近

Apr, 2016

Hardware-oriented Approximation of Convolutional Neural Networks

Philipp Gysel, Mohammad Motamedi, Soheil Ghiasi

TL;DR本文提出Ristretto，一种模型逼近框架，可以使用固定点算术和表示来压缩卷积和全连接层的权重和输出，并且可以通过微调将结果定制到具体的硬件设备，成功地将CaffeNet和SqueezeNet压缩到8位。

Abstract

High computational complexity hinders the widespread usage of convolutional neural networks (CNNs), especially in mobile devices. hardware accelerators are arguably the most promising approach for reducing both e