ID | 118049 |
著者 |
Nsinga, Robert
Tokushima University
|
キーワード | embedded system
IEEE 754-2008 floating-point
digital signal processor
Q format notation
|
資料タイプ |
学術雑誌論文
|
抄録 | Using less electric power or speeding up processing is catching the interests of researchers in deep learning. Quantization has offered distillation mechanisms that substitute floating numbers for integers, but little has been suggested about the floating numbers themselves. The use of Q-format notation reduces computational overheads that frees resources for the introduction of more operations. Our experiments, conditioned on varying regimes, introduce automatic differentiation on algorithms like the fast Fourier transforms and Winograd minimal filtering to reduce computational complexity (expressed in total number of MACs) and suggest a path towards the assistive intelligence concept. Empirical results show that, under specific heuristics, the Q-format number notation can overcome the shortfalls of floating numbers, especially for embedded systems. Further benchmarks like the FPBench standard give more details by comparing our proposals with common deep learning operations.
|
掲載誌名 |
Journal of Signal Processing
|
ISSN | 18801013
|
出版者 | Research Institute of Signal Processing
|
巻 | 26
|
号 | 5
|
開始ページ | 131
|
終了ページ | 140
|
発行日 | 2022-09-01
|
備考 | 利用は著作権の範囲内に限られる。
|
EDB ID | |
出版社版DOI | |
出版社版URL | |
フルテキストファイル | |
言語 |
eng
|
著者版フラグ |
出版社版
|
部局 |
理工学系
|