论文标题
可变形的蝴蝶：高度结构和稀疏的线性变换
Deformable Butterfly: A Highly Structured and Sparse Linear Transform
论文作者
论文摘要
我们引入了一种新型的线性变换，称为可变形蝴蝶（首次亮相），该变换概括了传统的蝴蝶矩阵，并且可以适应各种输入输出尺寸。它继承了传统蝴蝶的细粒元素可学习的层次结构，当部署到神经网络中时，首次亮相层中突出的结构和稀疏是一种用于网络压缩的新方法。我们将首次亮相作为完全连接和卷积层的标准替换，并证明了其在匀浆神经网络中的优越性，并在不损害准确性的情况下使其具有良好的特性，例如轻质重量和低推理复杂性。首次亮相层的无数变形引起的自然复杂性 - 准确性的权衡也为分析和实践研究开辟了新房间。代码和附录可在以下网址公开获取：https：//github.com/ruilin0212/debut。
We introduce a new kind of linear transform named Deformable Butterfly (DeBut) that generalizes the conventional butterfly matrices and can be adapted to various input-output dimensions. It inherits the fine-to-coarse-grained learnable hierarchy of traditional butterflies and when deployed to neural networks, the prominent structures and sparsity in a DeBut layer constitutes a new way for network compression. We apply DeBut as a drop-in replacement of standard fully connected and convolutional layers, and demonstrate its superiority in homogenizing a neural network and rendering it favorable properties such as light weight and low inference complexity, without compromising accuracy. The natural complexity-accuracy tradeoff arising from the myriad deformations of a DeBut layer also opens up new rooms for analytical and practical research. The codes and Appendix are publicly available at: https://github.com/ruilin0212/DeBut.
