TY - GEN
T1 - 2n RRR
T2 - 2018 New Generation of CAS, NGCAS 2018
AU - Ishikawa, Ryota
AU - Tawada, Masashi
AU - Yanagisawa, Masao
AU - Togawa, Nozomu
N1 - Funding Information:
ACKNOWLEDGMENT This study was supported in part by the MIC/SCOPE #171503005.
Publisher Copyright:
© 2018 IEEE.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2018/12/10
Y1 - 2018/12/10
N2 - In the fields of machine learning and image processing, cost-less circuits with low energy are required instead of extreme precision, and stochastic computing (SC), a type of approximate computing, is attracting attention. In SC, stochastic numbers (SNs), bit streams with values of the appearance rates of 1's, are used. SC enables calculations with simple circuits. To make the calculation results correct, duplication of an SN (gener-ating an SN with the same value) is required when using the SN with the same value. The conventional SN duplicator composed of a flip-flop (FF) has a problem that the output SN only depends on the input SN. Therefore, if the FF-based duplicator is used in a circuit with re-convergence paths, the output SN becomes erroneous. This paper proposes an SN duplicator, 2 n RRR, that can output more independent output by its improved flexibility of bit re-arrangement. With this duplicator, the errors of the hyperbolic tangent function are reduced by up to 50% compared to the duplicator that we proposed previously. Also, up to more than 99.9% of the circuit area is reduced compared to the implementation of binary computing.
AB - In the fields of machine learning and image processing, cost-less circuits with low energy are required instead of extreme precision, and stochastic computing (SC), a type of approximate computing, is attracting attention. In SC, stochastic numbers (SNs), bit streams with values of the appearance rates of 1's, are used. SC enables calculations with simple circuits. To make the calculation results correct, duplication of an SN (gener-ating an SN with the same value) is required when using the SN with the same value. The conventional SN duplicator composed of a flip-flop (FF) has a problem that the output SN only depends on the input SN. Therefore, if the FF-based duplicator is used in a circuit with re-convergence paths, the output SN becomes erroneous. This paper proposes an SN duplicator, 2 n RRR, that can output more independent output by its improved flexibility of bit re-arrangement. With this duplicator, the errors of the hyperbolic tangent function are reduced by up to 50% compared to the duplicator that we proposed previously. Also, up to more than 99.9% of the circuit area is reduced compared to the implementation of binary computing.
KW - Bit re-arrangement
KW - Re-convergence path
KW - Stochastic computing
KW - Stochastic number
KW - Stochastic number duplicator
UR - http://www.scopus.com/inward/record.url?scp=85060231240&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060231240&partnerID=8YFLogxK
U2 - 10.1109/NGCAS.2018.8572289
DO - 10.1109/NGCAS.2018.8572289
M3 - Conference contribution
AN - SCOPUS:85060231240
T3 - 2018 New Generation of CAS, NGCAS 2018
SP - 182
EP - 185
BT - 2018 New Generation of CAS, NGCAS 2018
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 20 November 2018 through 23 November 2018
ER -