Improvement of Accuracy of Well-Known Convoluational Neural Networks by Efficient Hybrid Strategy

Convolutional neural networks have existed for many years, but recently they have been developed to a greater depth and width than ever before with the increase in the computing power of graphics processing units. Convolutional neural networks are widely used in a variety of artificial intelligence applications, including in manufacturing, agriculture, and medicine. The use of artificial intelligence in various industrial fields is expected to increase. However, improvements in network training efficiency have not resulted in a reciprocal improvement in computational power for identification applications. This paper proposes several types of neural networks that are based on well-known networks such as AlexNet, GoogleNet, and ResNet, whose characteristics have been captured and implemented in lower layer neural networks. From the experimental results, using these hybrid neural networks can bring improved accuracy, with well optimized computational time costs compared to networks that require a large amount of computation.

[1]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[2]  Francesco Visin,et al.  A guide to convolution arithmetic for deep learning , 2016, ArXiv.

[3]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[4]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Sergey Ioffe,et al.  Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning , 2016, AAAI.

[6]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Xiaogang Wang,et al.  Deep Learning Face Representation from Predicting 10,000 Classes , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.