arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.11946cs.LGcs.NEstat.ML
收藏

基于稀疏神经网络的计算:对未来硬件的启示

Computation on Sparse Neural Networks: an Inspiration for Future Hardware

Alibaba DAMO Academy(阿里巴巴达摩院)

Fei Sun, Minghai Qin, Tianyun Zhang, Liu Liu, Yen-Kuang Chen, Yuan Xie

摘要:

神经网络模型被广泛应用于解决计算机视觉、个性化推荐、自然语言处理等具有挑战性的问题。这些模型的计算量非常大,达到了现有服务器和物联网设备的硬件极限。因此,找到更好的模型架构,具有更少的计算量,同时最大限度地保持精度是一个热门的研究课题。在各种旨在降低计算复杂度的机制中,识别模型权重和激活中的零值以避免计算它们是一个有前途的方向。本文从稀疏算法、软件框架和硬件加速等方面综述了稀疏神经网络计算的研究现状。我们观察到,搜索稀疏结构可以是高质量的模型探索的一般方法,除了高效率的模型执行的策略。讨论了权参数个数和模型结构对模型精度的影响。相应的模型被称为分别位于权重主导和结构主导区域。我们表明,对于实际复杂的问题,它是更有益的搜索大和稀疏的模型中的重量占主导地位的地区。为了实现这一目标,需要新的方法来搜索适当的稀疏结构,并且需要开发新的稀疏训练硬件以促进稀疏模型的快速迭代。

英文摘要:

Neural network models are widely used in solving many challenging problems, such as computer vision, personalized recommendation, and natural language processing. Those models are very computationally intensive and reach the hardware limit of the existing server and IoT devices. Thus, finding better model architectures with much less amount of computation while maximally preserving the accuracy is a popular research topic. Among various mechanisms that aim to reduce the computation complexity, identifying the zero values in the model weights and in the activations to avoid computing them is a promising direction. In this paper, we summarize the current status of the research on the computation of sparse neural networks, from the perspective of the sparse algorithms, the software frameworks, and the hardware accelerations. We observe that the search for the sparse structure can be a general methodology for high-quality model explorations, in addition to a strategy for high-efficiency model execution. We discuss the model accuracy influenced by the number of weight parameters and the structure of the model. The corresponding models are called to be located in the weight dominated and structure dominated regions, respectively. We show that for practically complicated problems, it is more beneficial to search large and sparse models in the weight dominated region. In order to achieve the goal, new approaches are required to search for proper sparse structures, and new sparse training hardware needs to be developed to facilitate fast iterations of sparse models.

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。

↑