TY - GEN
T1 - A Kernel Decomposition Architecture for Binary-weight Convolutional Neural Networks
AU - Kim, Hyeonuk
AU - Sim, Jaehyeong
AU - Choi, Yeongjae
AU - Kim, Lee Sup
N1 - Funding Information:
National Research Foundation of Korea (NRF)
Publisher Copyright:
© 2017 ACM.
PY - 2017/6/18
Y1 - 2017/6/18
N2 - The binary-weight CNN is one of the most efficient solutions for mobile CNNs. However, a large number of operations are required to process each image. To reduce such a huge operation count, we propose an energy-efficient kernel decomposition architecture, based on the observation that a large number of operations are redundant. In this scheme, all kernels are decomposed into sub-kernels to expose the common parts. By skipping the redundant computations, the operation count for each image was consequently reduced by 47.7%. Furthermore, a low cost bit-width quantization technique was implemented by exploiting the relative scales of the feature data. Experimental results showed that the proposed architecture achieves a 22% energy reduction.
AB - The binary-weight CNN is one of the most efficient solutions for mobile CNNs. However, a large number of operations are required to process each image. To reduce such a huge operation count, we propose an energy-efficient kernel decomposition architecture, based on the observation that a large number of operations are redundant. In this scheme, all kernels are decomposed into sub-kernels to expose the common parts. By skipping the redundant computations, the operation count for each image was consequently reduced by 47.7%. Furthermore, a low cost bit-width quantization technique was implemented by exploiting the relative scales of the feature data. Experimental results showed that the proposed architecture achieves a 22% energy reduction.
UR - http://www.scopus.com/inward/record.url?scp=85023628968&partnerID=8YFLogxK
U2 - 10.1145/3061639.3062189
DO - 10.1145/3061639.3062189
M3 - Conference contribution
AN - SCOPUS:85023628968
T3 - Proceedings - Design Automation Conference
BT - Proceedings of the 54th Annual Design Automation Conference 2017, DAC 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 54th Annual Design Automation Conference, DAC 2017
Y2 - 18 June 2017 through 22 June 2017
ER -