BriefGPT.xyz
Apr, 2018
SparseNet: 一种用于图像分类的稀疏 DenseNet
SparseNet: A Sparse DenseNet for Image Classification
HTML
PDF
Wenqi Liu, Kun Zeng
TL;DR
本文提出了一种稀疏DenseNet的方法，可以使神经网络在更低的参数和计算代价下实现深度、宽度和连接的同时，在CIFAR10和SVHN数据集上比现有state-of-the-art模型的性能更优，并且相比原始DenseNet小2.6倍，速度快3.7倍。
Abstract
deep neural networks
have made remarkable progresses on various
computer vision
tasks. Recent works have shown that depth, width and shortcut connections of networks are all vital to their performances. In this p
→