Recently, stochastic computing based on stochastic numbers attracts attention as an effective computation method, which realizes arithmetic operations by simple logic circuits with a tolerance of bit errors. When we input two or more identical values to a stochastic circuit, we require to duplicate a stochastic number. However, if bit streams of duplicated stochastic numbers are dependent on each other, their arithmetic operation results can be inaccurate. In this paper, we propose two stochastic number duplicators, called FSR and RRR. The stochastic numbers duplicated by the FSR and RRR duplicators have the equivalent values but have independent bit streams, effectively utilizing bit re-arrangement using randomized bit streams. Experimental evaluation results demonstrate that the RRR duplicator, in particular, obtains more accurate results even if a circuit has re-convergence paths, reducing the mean square errors by 20%-89% compared to a conventional stochastic number duplicator.
Ryota ISHIKAWA
Waseda University
Masashi TAWADA
Waseda University
Masao YANAGISAWA
Waseda University
Nozomu TOGAWA
Waseda University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Ryota ISHIKAWA, Masashi TAWADA, Masao YANAGISAWA, Nozomu TOGAWA, "Stochastic Number Duplicators Based on Bit Re-Arrangement Using Randomized Bit Streams" in IEICE TRANSACTIONS on Fundamentals,
vol. E101-A, no. 7, pp. 1002-1013, July 2018, doi: 10.1587/transfun.E101.A.1002.
Abstract: Recently, stochastic computing based on stochastic numbers attracts attention as an effective computation method, which realizes arithmetic operations by simple logic circuits with a tolerance of bit errors. When we input two or more identical values to a stochastic circuit, we require to duplicate a stochastic number. However, if bit streams of duplicated stochastic numbers are dependent on each other, their arithmetic operation results can be inaccurate. In this paper, we propose two stochastic number duplicators, called FSR and RRR. The stochastic numbers duplicated by the FSR and RRR duplicators have the equivalent values but have independent bit streams, effectively utilizing bit re-arrangement using randomized bit streams. Experimental evaluation results demonstrate that the RRR duplicator, in particular, obtains more accurate results even if a circuit has re-convergence paths, reducing the mean square errors by 20%-89% compared to a conventional stochastic number duplicator.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E101.A.1002/_p
Copy
@ARTICLE{e101-a_7_1002,
author={Ryota ISHIKAWA, Masashi TAWADA, Masao YANAGISAWA, Nozomu TOGAWA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Stochastic Number Duplicators Based on Bit Re-Arrangement Using Randomized Bit Streams},
year={2018},
volume={E101-A},
number={7},
pages={1002-1013},
abstract={Recently, stochastic computing based on stochastic numbers attracts attention as an effective computation method, which realizes arithmetic operations by simple logic circuits with a tolerance of bit errors. When we input two or more identical values to a stochastic circuit, we require to duplicate a stochastic number. However, if bit streams of duplicated stochastic numbers are dependent on each other, their arithmetic operation results can be inaccurate. In this paper, we propose two stochastic number duplicators, called FSR and RRR. The stochastic numbers duplicated by the FSR and RRR duplicators have the equivalent values but have independent bit streams, effectively utilizing bit re-arrangement using randomized bit streams. Experimental evaluation results demonstrate that the RRR duplicator, in particular, obtains more accurate results even if a circuit has re-convergence paths, reducing the mean square errors by 20%-89% compared to a conventional stochastic number duplicator.},
keywords={},
doi={10.1587/transfun.E101.A.1002},
ISSN={1745-1337},
month={July},}
Copy
TY - JOUR
TI - Stochastic Number Duplicators Based on Bit Re-Arrangement Using Randomized Bit Streams
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1002
EP - 1013
AU - Ryota ISHIKAWA
AU - Masashi TAWADA
AU - Masao YANAGISAWA
AU - Nozomu TOGAWA
PY - 2018
DO - 10.1587/transfun.E101.A.1002
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E101-A
IS - 7
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - July 2018
AB - Recently, stochastic computing based on stochastic numbers attracts attention as an effective computation method, which realizes arithmetic operations by simple logic circuits with a tolerance of bit errors. When we input two or more identical values to a stochastic circuit, we require to duplicate a stochastic number. However, if bit streams of duplicated stochastic numbers are dependent on each other, their arithmetic operation results can be inaccurate. In this paper, we propose two stochastic number duplicators, called FSR and RRR. The stochastic numbers duplicated by the FSR and RRR duplicators have the equivalent values but have independent bit streams, effectively utilizing bit re-arrangement using randomized bit streams. Experimental evaluation results demonstrate that the RRR duplicator, in particular, obtains more accurate results even if a circuit has re-convergence paths, reducing the mean square errors by 20%-89% compared to a conventional stochastic number duplicator.
ER -