arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.08596cs.LGstat.ML
收藏

卷积神经网络加速的研究进展

Recent Advances in Convolutional Neural Network Acceleration

National ASIC System Engineering Technology Research Center, Southeast University(东南大学国家专用集成电路系统工程技术研究中心);The Chinese University of Hong Kong(香港中文大学)

Qianru Zhang, Meng Zhang, Tinghuan Chen, Zhifei Sun, Yuzhe Ma, Bei Yu

摘要:

近年来,卷积神经网络(CNN)在诸如图像分类、模式识别和多媒体压缩的各种领域中已经显示出优异的性能。两个特征属性,局部连通性和权重共享,可以减少参数的数量,提高训练和推理过程中的处理速度。然而,随着数据的维度变得更高并且CNN架构变得更复杂,CNN的端到端方法或组合方式在计算上是密集的,这成为CNN的进一步实现的限制。因此,以更快的方式实现CNN是必要和紧迫的。在本文中,我们首先通过回顾各种研究论文,总结了有助于但不限于CNN的加速方法。我们提出了一个三个层次的分类法,即~加速方法的结构级、算法级和实现级。我们还分析了加速方法方面的CNN架构压缩,算法优化,和基于硬件的改进。最后,我们给出了这些加速和优化方法在每个层次上的不同角度进行了讨论。讨论表明,各层次的方法仍有较大的探索空间。通过整合如此广泛的学科,我们期望为对CNN加速感兴趣的研究人员提供全面的参考。

英文摘要:

In recent years, convolutional neural networks (CNNs) have shown great performance in various fields such as image classification, pattern recognition, and multi-media compression. Two of the feature properties, local connectivity and weight sharing, can reduce the number of parameters and increase processing speed during training and inference. However, as the dimension of data becomes higher and the CNN architecture becomes more complicated, the end-to-end approach or the combined manner of CNN is computationally intensive, which becomes limitation to CNN's further implementation. Therefore, it is necessary and urgent to implement CNN in a faster way. In this paper, we first summarize the acceleration methods that contribute to but not limited to CNN by reviewing a broad variety of research papers. We propose a taxonomy in terms of three levels, i.e.~structure level, algorithm level, and implementation level, for acceleration methods. We also analyze the acceleration methods in terms of CNN architecture compression, algorithm optimization, and hardware-based improvement. At last, we give a discussion on different perspectives of these acceleration and optimization methods within each level. The discussion shows that the methods in each level still have large exploration space. By incorporating such a wide range of disciplines, we expect to provide a comprehensive reference for researchers who are interested in CNN acceleration.

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。

↑