基于灵敏度分析的FPGM剪枝算法研究

针对等比例剪枝导致的重要卷积层剪枝过度、残留大量冗余参数以及精度损失较大的问题,在FPGM剪枝策略基础上融入灵敏度分析进行网络剪枝。算法采用精度反馈来分析每一层卷积层的重要性,控制单层剪枝比例分析每一层不同剪枝比例对精度损失的影响,获取各个卷积层的灵敏度;结合FPGM策略分析卷积层内卷积核的重要程度按灵敏度的剪枝比例剪掉不重要的卷积核,达到对神经网络进行有效剪枝的目的。实验结果表明,所提方法在MobileNet-v1和ResNet50上剪枝率为50%的情况下,精确度仅下降1.56%和0.11%;所提方法在精度损失一致下,ResNet50上具有更高剪枝率和更低计算量。

Research on FPGM pruning based on sensitivity analysis

The purpose is to deal with the problems of excessive pruning of important convolutional layers caused by equal-scale pruning, retain many redundant parameters and large loss of accuracy; This paper integrated sensitivity analysis on the basis of FPGM pruning strategy for network pruning. The algorithm used precision feedback to analyze the importance of each convolutional layer, control the single-layer pruning ratio, analyze the impact of different pruning ratios in each layer on accuracy loss, and obtained the sensitivity of each convolutional layer. It could be combined with FPGM to analyze the importance of the convolution kernel and cut the unimportant convolution kernel according to the pruning ratio of the sensitivity to complete the neural network pruning. The experimental results show that the accuracy of this method only drops by 1.56% and 0.11% when the pruning rate on MobileNet-v1 and ResNet50 is 50%. This method has a higher pruning rate and lower calculation amount on ResNet50 under the same accuracy loss.