Energy efficient CNN inference accelerator using fast fourier transform

Ya Chin Chung, Po Hsiang Cheng, Chih Wei Liu

    研究成果: Conference contribution同行評審

    2 引文 斯高帕斯(Scopus)

    摘要

    We use FFT-based convolution in frequency domain to reduce computational complexity in CNNs. The properties of conjugate symmetry and down-sampling is adopted to further reduce complexity. By eliminating filter weights in CNNs that can save computational requirement but lead to accuracy loss. The simulation result reveals that eliminating filter weights in frequency domain is more accurate than that in time domain. With the proposed design synthesized by TSMC 90 nm CMOS technology, the total latency, power and energy are considerably competitive. As a result, our FFT-based CNN accelerator is energy-efficient.

    原文English
    主出版物標題2019 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2019
    發行者Institute of Electrical and Electronics Engineers Inc.
    ISBN(電子)9781728106557
    DOIs
    出版狀態Published - 4月 2019
    事件2019 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2019 - Hsinchu, 台灣
    持續時間: 22 4月 201925 4月 2019

    出版系列

    名字2019 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2019

    Conference

    Conference2019 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2019
    國家/地區台灣
    城市Hsinchu
    期間22/04/1925/04/19

    指紋

    深入研究「Energy efficient CNN inference accelerator using fast fourier transform」主題。共同形成了獨特的指紋。

    引用此