Name
Title | ||
|---|---|---|
Energy efficient spiking neural network processing using approximate arithmetic units and variable precision weights |
Abstract | ||
|---|---|---|
•A field-programmable gate array (FPGA) based spiking neural network (SNN) accelerator architecture is proposed.•Approximate arithmetic units are utilized to realize energy efficient hardware implementation.•A variable precision method is proposed to minimize bit-width of weights.•The feasibility of utilizing the proposed method is verified via different SNN models. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1016/j.jpdc.2021.08.003 | Journal of Parallel and Distributed Computing |
Keywords | DocType | Volume |
Spiking neural network,Approximate computing,Field programmable gate array,Hardware accelerator | Journal | 158 |
ISSN | Citations | PageRank |
0743-7315 | 0 | 0.34 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yi Wang | 1 | 0 | 0.34 |
| Hao Zhang | 2 | 16 | 4.23 |
| Oh Kwang-il | 3 | 18 | 5.80 |
| Jae-Jin Lee | 4 | 0 | 0.34 |
| Seok-Bum Ko | 5 | 217 | 38.81 |