arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.13239cs.CV
收藏

利用非规则卷积核学习高效卷积网络

Learning Efficient Convolutional Networks through Irregular Convolutional Kernels

Central University of Finance and Economics(中央财经大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Watrix technology limited co. ltd(Watrix科技有限公司)

Weiyu Guo, Jiabin Ma, Liang Wang, Yongzhen Huang

摘要:

随着深度神经网络越来越多地用于适合于低功耗设备的应用中,一个基本的困境变得明显:趋势是增长模型以吸收增加的数据,这会导致内存密集型;然而,低功耗设备被设计为具有非常有限的内存,无法存储大型模型。参数修剪对于低功耗设备上的深度模型部署至关重要。现有的努力主要集中在设计高效的结构或修剪冗余连接的网络。它们通常对任务敏感,或者依赖于专用且昂贵的哈希存储策略。在这项工作中,我们介绍了一种新的方法来实现一个轻量级的模型,从重建卷积核的结构和有效的存储的观点。我们的方法将传统的平方卷积核转换为线段,并自动学习适当的策略来装备这些线段以模拟不同的特征。实验结果表明,我们的方法可以大量减少参数(在DenseNet-40上修剪69%)和计算(在DenseNet-40上修剪59%)的数量,同时保持可接受的性能(仅损失不到2%的准确度)。

英文摘要:

As deep neural networks are increasingly used in applications suited for low-power devices, a fundamental dilemma becomes apparent: the trend is to grow models to absorb increasing data that gives rise to memory intensive; however low-power devices are designed with very limited memory that can not store large models. Parameters pruning is critical for deep model deployment on low-power devices. Existing efforts mainly focus on designing highly efficient structures or pruning redundant connections for networks. They are usually sensitive to the tasks or relay on dedicated and expensive hashing storage strategies. In this work, we introduce a novel approach for achieving a lightweight model from the views of reconstructing the structure of convolutional kernels and efficient storage. Our approach transforms a traditional square convolution kernel to line segments, and automatically learn a proper strategy for equipping these line segments to model diverse features. The experimental results indicate that our approach can massively reduce the number of parameters (pruned 69% on DenseNet-40) and calculations (pruned 59% on DenseNet-40) while maintaining acceptable performance (only lose less than 2% accuracy).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑