publication . Conference object . Preprint . 2018

Accelerating Convolutional Neural Networks via Activation Map Compression

Georgiadis, Georgios;
Open Access
  • Published: 10 Dec 2018
  • Publisher: IEEE
Abstract
The deep learning revolution brought us an extensive array of neural network architectures that achieve state-of-the-art performance in a wide variety of Computer Vision tasks including among others, classification, detection and segmentation. In parallel, we have also been observing an unprecedented demand in computational and memory requirements, rendering the efficient use of neural networks in low-powered devices virtually unattainable. Towards this end, we propose a three-stage compression and acceleration pipeline that sparsifies, quantizes and entropy encodes activation maps of Convolutional Neural Networks. Sparsification increases the representational p...
Subjects
free text keywords: Entropy encoding, Computer science, Quantization (signal processing), Rendering (computer graphics), Convolutional neural network, Segmentation, Pattern recognition, Artificial intelligence, business.industry, business, Artificial neural network, Acceleration, Deep learning, Computer vision, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Related Organizations

[1] Zlib compressed data format specification version 3.3. https://tools.ietf.org/html/rfc1950. Accessed: 2018-05-17. 3

[2] J. Albericio, P. Judd, T. Hetherington, T. Aamodt, N. E. Jerger, and A. Moshovos. Cnvlutin: Ineffectual-neuron-free deep neural network computing. In 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture, 2016. 1

[3] M. Alwani, H. Chen, M. Ferdman, and P. Milder. Fused-layer cnn accelerators. In IEEE/ACM International Symposium on Microarchitecture, 2016. 3

[4] J. L. Ba, J. R. Kiros, and G. E. Hinton. Layer normalization. arXiv preprint arXiv:1607.06450, 2016. 3

[5] Z. Cai, X. He, J. Sun, and N. Vasconcelos. Deep learning with low precision by half-wave gaussian quantization. arXiv preprint arXiv:1702.00953, 2017. 5

Abstract
The deep learning revolution brought us an extensive array of neural network architectures that achieve state-of-the-art performance in a wide variety of Computer Vision tasks including among others, classification, detection and segmentation. In parallel, we have also been observing an unprecedented demand in computational and memory requirements, rendering the efficient use of neural networks in low-powered devices virtually unattainable. Towards this end, we propose a three-stage compression and acceleration pipeline that sparsifies, quantizes and entropy encodes activation maps of Convolutional Neural Networks. Sparsification increases the representational p...
Subjects
free text keywords: Entropy encoding, Computer science, Quantization (signal processing), Rendering (computer graphics), Convolutional neural network, Segmentation, Pattern recognition, Artificial intelligence, business.industry, business, Artificial neural network, Acceleration, Deep learning, Computer vision, Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Related Organizations

[1] Zlib compressed data format specification version 3.3. https://tools.ietf.org/html/rfc1950. Accessed: 2018-05-17. 3

[2] J. Albericio, P. Judd, T. Hetherington, T. Aamodt, N. E. Jerger, and A. Moshovos. Cnvlutin: Ineffectual-neuron-free deep neural network computing. In 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture, 2016. 1

[3] M. Alwani, H. Chen, M. Ferdman, and P. Milder. Fused-layer cnn accelerators. In IEEE/ACM International Symposium on Microarchitecture, 2016. 3

[4] J. L. Ba, J. R. Kiros, and G. E. Hinton. Layer normalization. arXiv preprint arXiv:1607.06450, 2016. 3

[5] Z. Cai, X. He, J. Sun, and N. Vasconcelos. Deep learning with low precision by half-wave gaussian quantization. arXiv preprint arXiv:1702.00953, 2017. 5

Powered by OpenAIRE Research Graph
Any information missing or wrong?Report an Issue