用固定卷积核训练CNN，仍能学出有效特征。
Convolutional Neural Networks Do Work with Pre-Defined Filters
- 所有卷积核预先设定，训练时不更新。
- 在多个数据集上达到接近标准CNN的准确率。
- 适合研究CNN内部信息处理机制的学者。
我们提出一类新型卷积神经网络——预定义滤波器卷积神经网络（PFCNN），其中所有n×n卷积核（n>1）在训练过程中保持预定义且恒定。该网络采用一种特殊深度可分离卷积结构，称为预定义滤波器模块（PFM）。通道卷积部分使用来自仅16种不同预定义核的1×n×n核；1×1卷积部分则学习这些预定义滤波器输出的线性组合。尽管存在严格限制，仍能提取复杂且具有判别性的特征。这一发现为理解深度CNN中信息处理方式提供了新视角。我们在Caltech101、CIFAR10、CUB-200-2011、FGVC-Aircraft、Flowers102和Stanford Cars等主流数据集上验证了PFCNN的有效性。代码已开源：https://github.com/Criscraft/PredefinedFilterNetworks。
原文摘要 · Abstract (English)
We present a novel class of Convolutional Neural Networks called Pre-defined Filter Convolutional Neural Networks (PFCNNs), where all nxn convolution kernels with n>1 are pre-defined and constant during training. It involves a special form of depthwise convolution operation called a Pre-defined Filter Module (PFM). In the channel-wise convolution part, the 1xnxn kernels are drawn from a fixed pool of only a few (16) different pre-defined kernels. In the 1x1 convolution part linear combinations of the pre-defined filter outputs are learned. Despite this harsh restriction, complex and discriminative features are learned. These findings provide a novel perspective on the way how information is processed within deep CNNs. We discuss various properties of PFCNNs and prove their effectiveness using the popular datasets Caltech101, CIFAR10, CUB-200-2011, FGVC-Aircraft, Flowers102, and Stanford Cars. Our implementation of PFCNNs is provided on Github https://github.com/Criscraft/PredefinedFilterNetworks
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。