An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits

    研究成果: Conference contribution

    抄録

    Due to the continuous demand of small-sized cir- cuits in the fields of image processing and artificial intelligence, stochastic computing has attracted much attention. In stochastic computing, stochastic numbers composed of randomly generated bit streams are used for computation. When inputting two or more identical values to a stochastic circuit, their arithmetic operation results can be inaccurate. In this paper, a stochastic number duplicator called RRR (Register based Re-arrangement circuit using a Random bit stream) duplicator is introduced and is evaluated by applying to composite arithmetic circuits including re-convergence paths. The stochastic numbers duplicated by RRR duplicator have the equivalent values to its input stochastic numbers but have independent bit streams, effectively utilizing bit re-arrangement based on randomized bit streams. It is applied to composite arithmetic circuits and its accuracy and area/delay requirements are evaluated. The results demonstrate that RRR duplicator obtains more accurate results in a circuit with re- convergence paths, reducing the mean square errors by 64%- 94% compared to a conventional stochastic number duplicator.

    元の言語English
    ホスト出版物のタイトル2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018
    編集者Mihalis Maniatakos, Dan Alexandrescu, Dimitris Gizopoulos, Panagiota Papavramidou
    出版者Institute of Electrical and Electronics Engineers Inc.
    ページ53-56
    ページ数4
    ISBN(電子版)9781538659922
    DOI
    出版物ステータスPublished - 2018 9 26
    イベント24th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2018 - Platja D'Aro, Spain
    継続期間: 2018 7 22018 7 4

    Other

    Other24th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2018
    Spain
    Platja D'Aro
    期間18/7/218/7/4

    Fingerprint

    Networks (circuits)
    Composite materials
    Mean square error
    Artificial intelligence
    Image processing

    ASJC Scopus subject areas

    • Hardware and Architecture
    • Software
    • Safety, Risk, Reliability and Quality

    これを引用

    Ishikawa, R., Tawada, M., Yanagisawa, M., & Togawa, N. (2018). An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits. : M. Maniatakos, D. Alexandrescu, D. Gizopoulos, & P. Papavramidou (版), 2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018 (pp. 53-56). [8474263] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IOLTS.2018.8474263

    An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits. / Ishikawa, Ryota; Tawada, Masashi; Yanagisawa, Masao; Togawa, Nozomu.

    2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018. 版 / Mihalis Maniatakos; Dan Alexandrescu; Dimitris Gizopoulos; Panagiota Papavramidou. Institute of Electrical and Electronics Engineers Inc., 2018. p. 53-56 8474263.

    研究成果: Conference contribution

    Ishikawa, R, Tawada, M, Yanagisawa, M & Togawa, N 2018, An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits. : M Maniatakos, D Alexandrescu, D Gizopoulos & P Papavramidou (版), 2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018., 8474263, Institute of Electrical and Electronics Engineers Inc., pp. 53-56, 24th IEEE International Symposium on On-Line Testing and Robust System Design, IOLTS 2018, Platja D'Aro, Spain, 18/7/2. https://doi.org/10.1109/IOLTS.2018.8474263
    Ishikawa R, Tawada M, Yanagisawa M, Togawa N. An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits. : Maniatakos M, Alexandrescu D, Gizopoulos D, Papavramidou P, 編集者, 2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 53-56. 8474263 https://doi.org/10.1109/IOLTS.2018.8474263
    Ishikawa, Ryota ; Tawada, Masashi ; Yanagisawa, Masao ; Togawa, Nozomu. / An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits. 2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018. 編集者 / Mihalis Maniatakos ; Dan Alexandrescu ; Dimitris Gizopoulos ; Panagiota Papavramidou. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 53-56
    @inproceedings{bf6968cb388d4b72956b6555769e0d76,
    title = "An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits",
    abstract = "Due to the continuous demand of small-sized cir- cuits in the fields of image processing and artificial intelligence, stochastic computing has attracted much attention. In stochastic computing, stochastic numbers composed of randomly generated bit streams are used for computation. When inputting two or more identical values to a stochastic circuit, their arithmetic operation results can be inaccurate. In this paper, a stochastic number duplicator called RRR (Register based Re-arrangement circuit using a Random bit stream) duplicator is introduced and is evaluated by applying to composite arithmetic circuits including re-convergence paths. The stochastic numbers duplicated by RRR duplicator have the equivalent values to its input stochastic numbers but have independent bit streams, effectively utilizing bit re-arrangement based on randomized bit streams. It is applied to composite arithmetic circuits and its accuracy and area/delay requirements are evaluated. The results demonstrate that RRR duplicator obtains more accurate results in a circuit with re- convergence paths, reducing the mean square errors by 64{\%}- 94{\%} compared to a conventional stochastic number duplicator.",
    keywords = "bit re-arrangement, re-convergence path, stochastic computing, stochastic number, stochastic numbrestochastic number duplicator",
    author = "Ryota Ishikawa and Masashi Tawada and Masao Yanagisawa and Nozomu Togawa",
    year = "2018",
    month = "9",
    day = "26",
    doi = "10.1109/IOLTS.2018.8474263",
    language = "English",
    pages = "53--56",
    editor = "Mihalis Maniatakos and Dan Alexandrescu and Dimitris Gizopoulos and Panagiota Papavramidou",
    booktitle = "2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - GEN

    T1 - An Effective Stochastic Number Duplicator and Its Evaluations Using Composite Arithmetic Circuits

    AU - Ishikawa, Ryota

    AU - Tawada, Masashi

    AU - Yanagisawa, Masao

    AU - Togawa, Nozomu

    PY - 2018/9/26

    Y1 - 2018/9/26

    N2 - Due to the continuous demand of small-sized cir- cuits in the fields of image processing and artificial intelligence, stochastic computing has attracted much attention. In stochastic computing, stochastic numbers composed of randomly generated bit streams are used for computation. When inputting two or more identical values to a stochastic circuit, their arithmetic operation results can be inaccurate. In this paper, a stochastic number duplicator called RRR (Register based Re-arrangement circuit using a Random bit stream) duplicator is introduced and is evaluated by applying to composite arithmetic circuits including re-convergence paths. The stochastic numbers duplicated by RRR duplicator have the equivalent values to its input stochastic numbers but have independent bit streams, effectively utilizing bit re-arrangement based on randomized bit streams. It is applied to composite arithmetic circuits and its accuracy and area/delay requirements are evaluated. The results demonstrate that RRR duplicator obtains more accurate results in a circuit with re- convergence paths, reducing the mean square errors by 64%- 94% compared to a conventional stochastic number duplicator.

    AB - Due to the continuous demand of small-sized cir- cuits in the fields of image processing and artificial intelligence, stochastic computing has attracted much attention. In stochastic computing, stochastic numbers composed of randomly generated bit streams are used for computation. When inputting two or more identical values to a stochastic circuit, their arithmetic operation results can be inaccurate. In this paper, a stochastic number duplicator called RRR (Register based Re-arrangement circuit using a Random bit stream) duplicator is introduced and is evaluated by applying to composite arithmetic circuits including re-convergence paths. The stochastic numbers duplicated by RRR duplicator have the equivalent values to its input stochastic numbers but have independent bit streams, effectively utilizing bit re-arrangement based on randomized bit streams. It is applied to composite arithmetic circuits and its accuracy and area/delay requirements are evaluated. The results demonstrate that RRR duplicator obtains more accurate results in a circuit with re- convergence paths, reducing the mean square errors by 64%- 94% compared to a conventional stochastic number duplicator.

    KW - bit re-arrangement

    KW - re-convergence path

    KW - stochastic computing

    KW - stochastic number

    KW - stochastic numbrestochastic number duplicator

    UR - http://www.scopus.com/inward/record.url?scp=85055817128&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85055817128&partnerID=8YFLogxK

    U2 - 10.1109/IOLTS.2018.8474263

    DO - 10.1109/IOLTS.2018.8474263

    M3 - Conference contribution

    AN - SCOPUS:85055817128

    SP - 53

    EP - 56

    BT - 2018 IEEE 24th International Symposium on On-Line Testing and Robust System Design, IOLTS 2018

    A2 - Maniatakos, Mihalis

    A2 - Alexandrescu, Dan

    A2 - Gizopoulos, Dimitris

    A2 - Papavramidou, Panagiota

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -