ABSTRACT
The training process of a deep neural network commonly consists of three phases: forward propagation, backward propagation, and weight update. In this paper, we propose a hardware architecture to accelerate the backward propagation. Our approach applies to neural networks that use rectified linear unit. Considering that the backward propagation results in a zero activation gradient when the corresponding activation is zero, we can safely skip the gradient calculation. Based on this observation, we design an efficient hardware accelerator for training deep neural networks by selectively computing gradients. We show the effectiveness of our approach through experiments with various network models.
- Jorge Albericio, Patrick Judd, Tayler Hetherington, Tor Aamodt, Natalie Enright Jerger, and Andreas Moshovos. 2016. Cnvlutin: Ineffectual-neuron-free deep neural network computing. In ACM SIGARCH Computer Architecture News, Vol. 44. IEEE Press, 1--13. Google ScholarDigital Library
- Tianshi Chen, Zidong Du, Ninghui Sun, Jia Wang, Chengyong Wu, Yunji Chen, and Olivier Temam. 2014. Diannao: A small-footprint high-throughput accelerator for ubiquitous machine-learning. ACM Sigplan Notices 49, 4 (2014), 269--284. Google ScholarDigital Library
- Yunji Chen, Tao Luo, Shaoli Liu, Shijin Zhang, Liqiang He, Jia Wang, Ling Li, Tianshi Chen, Zhiwei Xu, Ninghui Sun, et al. 2014. Dadiannao: A machine-learning supercomputer. In Proceedings of the 47th Annual IEEE/ACM International Symposium on Microarchitecture. IEEE Computer Society, 609--622. Google ScholarDigital Library
- Yu-Hsin Chen, Tushar Krishna, Joel S Emer, and Vivienne Sze. 2017. Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks. IEEE Journal of Solid-State Circuits 52, 1 (2017), 127--138.Google ScholarCross Ref
- Awni Hannun, Carl Case, Jared Casper, Bryan Catanzaro, Greg Diamos, Erich Elsen, Ryan Prenger, Sanjeev Satheesh, Shubho Sengupta, Adam Coates, et al. 2014. Deep speech: Scaling up end-to-end speech recognition. arXiv preprint arXiv:1412.5567 (2014).Google Scholar
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 770--778.Google ScholarCross Ref
- Sergey Ioffe and Christian Szegedy. 2015. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In International Conference on Machine Learning. 448--456. Google ScholarDigital Library
- Dongyoung Kim, Junwhan Ahn, and Sungjoo Yoo. 2018. ZeNA: Zero-Aware Neural Network Accelerator. IEEE Design & Test 35, 1 (2018), 39--46.Google ScholarCross Ref
- Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. 2012. Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems. 1097--1105. Google ScholarDigital Library
- Vinod Nair and Geoffrey E Hinton. 2010. Rectified linear units improve restricted boltzmann machines. In Proceedings of the 27th international conference on machine learning (ICML-10). 807--814. Google ScholarDigital Library
- Minsoo Rhu, Mike O'Connor, Niladrish Chatterjee, Jeff Pool, Youngeun Kwon, and Stephen W Keckler. 2018. Compressing DMA engine: Leveraging activation sparsity for training deep neural networks. In 2018 IEEE International Symposium on High Performance Computer Architecture (HPCA). IEEE, 78--91.Google ScholarCross Ref
- Paul Rosenfeld, Elliott Cooper-Balis, and Bruce Jacob. 2011. DRAMSim2: A cycle accurate memory system simulator. IEEE Computer Architecture Letters 10, 1 (2011), 16--19. Google ScholarDigital Library
- Karen Simonyan and Andrew Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014).Google Scholar
- Andrea Vedaldi and Karel Lenc. 2015. Matconvnet: Convolutional neural networks for matlab. In Proceedings of the 23rd ACM international conference on Multimedia. ACM, 689--692. Google ScholarDigital Library
- Tom Young, Devamanyu Hazarika, Soujanya Poria, and Erik Cambria. 2018. Recent trends in deep learning based natural language processing. ieee Computational intelligence magazine 13, 3 (2018), 55--75.Google Scholar
- Shijin Zhang, Zidong Du, Lei Zhang, Huiying Lan, Shaoli Liu, Ling Li, Qi Guo, Tianshi Chen, and Yunji Chen. 2016. Cambricon-x: An accelerator for sparse neural networks. In The 49th Annual IEEE/ACM International Symposium on Microarchitecture. IEEE Press, 20. Google ScholarDigital Library
Index Terms
- Acceleration of DNN Backward Propagation by Selective Computation of Gradients
Recommendations
TJU-DNN: A trajectory-unified framework for training deep neural networks and its applications
AbstractThe training method for deep neural networks mainly adopts the gradient descent (GD) method. These methods, however, are very sensitive to initialization and hyperparameters. In this paper, an enhanced gradient descent method guided by ...
A novel softplus linear unit for deep convolutional neural networks
Current improvements in the performance of deep neural networks are partly due to the proposition of rectified linear units. A ReLU activation function outputs zero for negative component, inducing the death of some neurons and a bias shift of the ...
Steepest descent using smoothed gradients
This paper reviews solving differential equations using the least-squares steepest descent method with Sobolev gradients. The method's superiority over standard steepest descent with a Euclidean gradient is explained in terms of stability and the ...
Comments