论文标题
有效的CNN体系结构设计以可视化为指导
Efficient CNN Architecture Design Guided by Visualization
论文作者
论文摘要
现代有效的卷积神经网络（CNN）始终使用深度可分离的卷积（DSC）和神经体系结构搜索（NAS）来减少参数数量和计算复杂性。但是网络的一些固有特征被忽略了。受到可视化功能地图和n $ \ times $ n（n $> $ 1）卷积内核的启发，本文介绍了几种准则，以进一步提高参数效率和推理速度。基于这些准则，我们的参数有效的CNN体系结构称为\ textit {vgnetg}，比以前的网络比以前的网络获得了更好的准确性和更低的延迟，降低了约30％$ \厚度$ 50％的参数。我们的VGNETG-1.0MP在ImagEnet分类数据集上具有99万参数的67.7％TOP-1精度，而69.2％的TOP-1精度具有114m参数。 此外，我们证明边缘检测器可以通过用固定的边缘检测核代替N $ \ times $ n内核来替换可学习的深度卷积层以混合特征。我们的VGNETF-1.5MP存档64.4％（ - 3.2％）的前1位准确性和66.2％（-1.4％）的TOP-1准确度具有额外的高斯内核。
Modern efficient Convolutional Neural Networks(CNNs) always use Depthwise Separable Convolutions(DSCs) and Neural Architecture Search(NAS) to reduce the number of parameters and the computational complexity. But some inherent characteristics of networks are overlooked. Inspired by visualizing feature maps and N$\times$N(N$>$1) convolution kernels, several guidelines are introduced in this paper to further improve parameter efficiency and inference speed. Based on these guidelines, our parameter-efficient CNN architecture, called \textit{VGNetG}, achieves better accuracy and lower latency than previous networks with about 30%$\thicksim$50% parameters reduction. Our VGNetG-1.0MP achieves 67.7% top-1 accuracy with 0.99M parameters and 69.2% top-1 accuracy with 1.14M parameters on ImageNet classification dataset. Furthermore, we demonstrate that edge detectors can replace learnable depthwise convolution layers to mix features by replacing the N$\times$N kernels with fixed edge detection kernels. And our VGNetF-1.5MP archives 64.4%(-3.2%) top-1 accuracy and 66.2%(-1.4%) top-1 accuracy with additional Gaussian kernels.
