Optimizing Performance of Convolutional Neural Network Using Computing Technique

Based on the deep learning algorithms, there are many recent rapid growths of applications. A deep learning algorithm which is expanded from Artificial Neural Networks, and it is extensively used for picture categorization and identification. Still there is no Neural Network based computing technique, i.e. pipelining has been signified in order to assist the existence of deep neural network in terms of accuracy. In this paper, Computing-based convolutional neural network system produced highly precise and productive system is used. The proposed convolutional neural network will compare with previous convolutional neural network executed on an FPGA using a computing technique to optimize the time delay and power consumption of the system, with high accuracy as compared to previous conventional neural network implementations.

[1]  Y. LeCun,et al.  Learning methods for generic object recognition with invariance to pose and lighting , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[2]  Yoshua Bengio,et al.  Convolutional networks for images, speech, and time series , 1998 .

[3]  Jason Cong,et al.  Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks , 2015, FPGA.

[4]  Kiyoung Choi,et al.  FPGA implementation of convolutional neural network based on stochastic computing , 2017, 2017 International Conference on Field Programmable Technology (ICFPT).

[5]  Song Han,et al.  EIE: Efficient Inference Engine on Compressed Deep Neural Network , 2016, 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA).

[6]  Natalie D. Enright Jerger,et al.  Cnvlutin: Ineffectual-Neuron-Free Deep Neural Network Computing , 2016, 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA).

[7]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[8]  Mingjie Lin,et al.  Stochastic-based multi-stage streaming realization of deep convolutional neural network , 2017, 2017 18th International Symposium on Quality Electronic Design (ISQED).

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

[10]  Mitra Mirhassani,et al.  Efficient VLSI Implementation of Neural Networks With Hyperbolic Tangent Activation Function , 2014, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[11]  David A. Patterson,et al.  In-datacenter performance analysis of a tensor processing unit , 2017, 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA).

[12]  Jongeun Lee,et al.  A new stochastic computing multiplier with application to deep convolutional neural networks , 2017, 2017 54th ACM/EDAC/IEEE Design Automation Conference (DAC).