Saffar Mohsen, Kalhor Ahmad, Habibnia Ali
School of Electrical and Computer Engineering, College of Engineering, University of Tehran, Tehran, Iran.
Department of Economics and the Computational Modeling and Data Analytics, College of Science, Virginia Polytechnic Institute and State University, Blacksburg, USA.
Sci Rep. 2024 Jul 30;14(1):17590. doi: 10.1038/s41598-024-68172-6.
This paper proposes a geometric-based technique for compressing convolutional neural networks to accelerate computations and improve generalization by eliminating non-informative components. The technique utilizes a geometric index called separation index to evaluate the functionality of network elements such as layers and filters. By applying this index along with center-based separation index, a systematic algorithm is proposed that optimally compresses convolutional and fully connected layers. The algorithm excludes layers with low performance, selects the best subset of filters in the filtering layers, and tunes the parameters of fully connected layers using center-based separation index. An illustrative example of classifying CIFAR-10 dataset is presented to explain the algorithm step-by-step. The proposed method achieves impressive pruning results on networks trained by CIFAR-10 and ImageNet datasets, with 87.5%, 77.6%, and 78.8% of VGG16, GoogLeNet, and DenseNet parameters pruned, respectively. Comparisons with state-of-the-art works are provided to demonstrate the effectiveness of the proposed method.
本文提出了一种基于几何的技术，用于压缩卷积神经网络，通过消除无信息成分来加速计算并提高泛化能力。该技术利用一种称为分离指数的几何指标来评估诸如层和滤波器等网络元素的功能。通过应用此指标以及基于中心的分离指数，提出了一种系统算法，该算法可对卷积层和全连接层进行最优压缩。该算法排除性能较低的层，在滤波层中选择最佳滤波器子集，并使用基于中心的分离指数调整全连接层的参数。给出了一个对CIFAR-10数据集进行分类的示例，逐步解释该算法。所提出的方法在由CIFAR-10和ImageNet数据集训练的网络上取得了令人印象深刻的剪枝结果，分别对VGG16、GoogLeNet和DenseNet的87.5%、77.6%和78.8%的参数进行了剪枝。与现有技术进行了比较，以证明所提出方法的有效性。
