摘要
针对基于权重归一化方法的卷积神经网络(CNN)转换方法应用于事件流数据时准确率损失较大以及浮点网络难以在硬件上高效部署等问题,提出一种面向动态事件流的网络转换方法。首先,重构事件流数据并输入CNN进行训练,在训练过程中采用量化激活函数降低转换的准确率损失,并使用对称定点量化方法以减少参数存储量;其次,在网络转换中采用脉冲计数等价原理而非频率等价原理以更好地适应数据的稀疏性。实验结果表明,与使用传统激活函数相比,采用量化激活函数的脉冲卷积神经网络(SCNN)在N-MNIST、POKER-DVS和MNIST-DVS这三个动态事件流数据集上的识别准确率分别提高了0.29个百分点、8.52个百分点和3.95个百分点,转换损失分别降低了21.77%、100.00%和92.48%;此外,相较于基于权重归一化方法生成的高精度SCNN,所提量化SCNN在识别准确率相当的情况下可以有效节省约75%的存储空间,并且在N-MNIST和MNIST-DVS数据集上的转换损失分别降低了6.79%和46.29%。
Since Convolutional Neural Network(CNN) conversion method based on the weight normalization method for event stream data has a large loss of accuracy and the effective deployment of floating-point networks is difficult on hardware, a network conversion method for dynamic event stream was proposed. Firstly, the event stream data was reconstructed as the input of CNN for training. In the training process, the quantized activation function was adopted to reduce the accuracy loss, and a symmetric fixed-point quantization method was used to reduce the parameter storage. Then, instead of equivalence principle, pulse count equivalence principle was used to adapt to the sparsity of data better. Experimental results show that on three datasets N-MNIST, POKER-DVS and MNIST-DVS, compared with using the traditional activation function, Spiking Convolutional Neural Network(SCNN) using the quantized activation function has the recognition accuracy improved by 0. 29 percentage points, 8. 52 percentage points and 3. 95 percentage points respectively, and the conversion loss reduced by 21. 77%, 100. 00% and 92. 48% respectively. Meanwhile, the proposed quantized SCNN can effectively save 75% of storage space compared with high-precision SCNN generated on the basis of the weight normalization method, and has the conversion loss on N-MNIST and MNIST-DVS datasets reduced by 6. 79% and 46. 29% respectively.
作者
张宇豪
袁孟雯
陆宇婧
燕锐
唐华锦
ZHANG Yuhao;YUAN Mengwen;LU Yujing;YAN Rui;TANG Huajin(College of Computer Science,Sichuan University,Chengdu Sichuan 610065,China;Research Center for Intelligent Computing Hardware,Zhejiang Laboratory,Hangzhou Zhejiang 311100,China;College of Computer Science and Technology,Zhejiang University of Technology,Hangzhou Zhejiang 310023,China;College of Computer Science and Technology,Zhejiang University,Hangzhou Zhejiang 310027,China)
出处
《计算机应用》
CSCD
北大核心
2022年第10期3033-3039,共7页
journal of Computer Applications
基金
国家自然科学基金委员会-中国工程物理研究院“NSAF”联合基金资助项目(U2030204)
国家自然科学基金资助项目(61773271)
之江实验室科研攻关项目(2021KC0AC01)。
关键词
神经网络转换
动态事件流
量化激活函数
定点量化
脉冲卷积神经网络
neural network conversion
dynamic event stream
quantized activation function
fixed-point quantization
Spiking Convolutional Neural Network(SCNN)
