摘要
卷积神经网络(convolutional neural network,CNN)已被广泛应用于图像识别领域,其自身的超参数对图像分类问题中分类错误率的大小有较大的影响。为进一步优化CNN超参数,提出了基于Softmax回归的相对概率变化比。应用相对概率变化比寻找对图像分类影响较大的超参数,并根据超参数的重要性大小依次对其进行调整。为验证相对概率比的有效性,在网络架构1和网络架构2下进行了调参实验。实验结果表明,Softmax回归的相对概率变化比在不同的网络架构下均可反映CNN超参数对分类错误率的影响,且有助于找到一个更优的超参数组合,降低分类错误率。在MNIST和CIFAR-10数据集上的对比实验表明,研究结果在不同数据集下都适用。
Convolutional neural network(CNN)has been widely used in the field of image classification.Its hyper-parameters have a great impact on the misclassification rate.In order to further optimize CNN hyper-parameters,the change ratio of relative probability based on Softmax regression was introduced.The change ratio of relative probability was used to find the hyper-parameters which have great influence on the image classification problem,and the hyper-parameters were adjusted according to their importance.In order to verify the effectiveness of the change ratio of relative probability,experiments were carried out in Architecture 1 and Architecture 2.The experiment results show that the concept of the change ratio of relative probability introduced in Softmax regression can reflect the effect of CNN hyper-parameters on misclassification rate in both architectures,and it is more helpful to find a better combination of CNN hyper-parameters and reduce the misclassification rate.The comparative experiments on MNIST and CIFAR-10 datasets show that the above conclusions are applicable to different datasets.
作者
李慧
周溪召
施柏州
LI Hui;ZHOU Xizhao;SHI Baizhou(Business School,University of Shanghai for Science and Technology,Shanghai 200093,China;College of Science and Engineering,Chaoyang University of Technology,Taichung 413310,China)
出处
《上海理工大学学报》
CAS
CSCD
北大核心
2021年第3期219-226,共8页
Journal of University of Shanghai For Science and Technology
基金
国家自然科学基金资助项目(61273042)。
关键词
卷积神经网络
超参数组合
Softmax回归
相对概率变化比
convolutional neural network
combination of hyper-parameters
Softmax regression
change ratio of relative probability
