当前位置: X-MOL 学术Neural Netw. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Self-grouping convolutional neural networks
Neural Networks ( IF 6.0 ) Pub Date : 2020-09-17 , DOI: 10.1016/j.neunet.2020.09.015
Qingbei Guo , Xiao-Jun Wu , Josef Kittler , Zhiquan Feng

Although group convolution operators are increasingly used in deep convolutional neural networks to improve the computational efficiency and to reduce the number of parameters, most existing methods construct their group convolution architectures by a predefined partitioning of the filters of each convolutional layer into multiple regular filter groups with an equal spatial group size and data-independence, which prevents a full exploitation of their potential. To tackle this issue, we propose a novel method of designing self-grouping convolutional neural networks, called SG-CNN, in which the filters of each convolutional layer group themselves based on the similarity of their importance vectors. Concretely, for each filter, we first evaluate the importance value of their input channels to identify the importance vectors, and then group these vectors by clustering. Using the resulting data-dependent centroids, we prune the less important connections, which implicitly minimizes the accuracy loss of the pruning, thus yielding a set of diverse group convolution filters. Subsequently, we develop two fine-tuning schemes, i.e. (1) both local and global fine-tuning and (2) global only fine-tuning, which experimentally deliver comparable results, to recover the recognition capacity of the pruned network. Comprehensive experiments carried out on the CIFAR-10/100 and ImageNet datasets demonstrate that our self-grouping convolution method adapts to various state-of-the-art CNN architectures, such as ResNet and DenseNet, and delivers superior performance in terms of compression ratio, speedup and recognition accuracy. We demonstrate the ability of SG-CNN to generalize by transfer learning, including domain adaption and object detection, showing competitive results. Our source code is available at https://github.com/QingbeiGuo/SG-CNN.git.



中文翻译:

自分组卷积神经网络

尽管在深度卷积神经网络中越来越多地使用组卷积算子来提高计算效率并减少参数数量,但是大多数现有方法都是通过将每个卷积层的滤波器预定义为多个常规滤波器组来构造其组卷积体系结构的。相等的空间组大小和数据独立性,从而阻止了其潜力的充分利用。为了解决这个问题,我们提出了一种新的设计自分组卷积神经网络的方法,称为SG-CNN,其中每个卷积层的过滤器根据其重要性矢量的相似性将它们自身分组。具体而言,对于每个过滤器,我们首先评估其输入通道的重要性值以识别重要性向量,然后通过聚类对这些向量进行分组。使用结果依赖于数据的形心,我们会修剪次要的连接,从而隐式地减少了修剪的准确性损失,从而产生了一组不同的组卷积过滤器。随后,我们开发了两种微调方案,即(1)本地和全局微调,以及(2)全局仅微调,它们在实验上提供了可比的结果,以恢复修剪后的网络的识别能力。在CIFAR-10 / 100和ImageNet数据集上进行的综合实验表明,我们的自分组卷积方法适用于各种最新的CNN架构,例如ResNet和DenseNet,并且在压缩率方面具有出色的性能。 ,加速和识别准确性。我们展示了SG-CNN通过迁移学习进行泛化的能力,包括域自适应和对象检测,从而显示出具有竞争力的结果。我们的源代码位于https://github.com/QingbeiGuo/SG-CNN.git。

更新日期:2020-10-11
down
wechat
bug