.png){kind=icon}
.png){kind=icon}
Return
Energy-Efficient Convolution Architecture Based on Rescheduled Dataflow
.png){kind=icon}
PRE
AI
DOI：10.1109/TCSI.2018.2840092.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper presents a rescheduled dataflow of convolution and its hardware architecture that can enhance energy efficiency. For convolution involving a large amount of computations and memory accesses, previous accelerators employed parallel processing elements to meet real-time constraints. Though the previous approaches made a success in implementing complex convolution models, they load the same input features and filter weights from on-chip memories multiple times due to the iterative property of convolution operations, suffering from high energy consumption. To mitigate redundant memory accesses, a novel dataflow is proposed that computes convolution operations incrementally so as to reuse the loaded data as maximally as possible. In addition, several convolution accelerators supporting the rescheduled dataflow are investigated, and qualitative and quantitative analyses are performed to suggest a promising candidate for various convolution models. Simulation results show that the energy efficiency of the proposed accelerator outperforms that of the previous accelerator significantly.
Keywords:
Convolution
deep neural networks
convolutional neural networks
energy-efficient accelerator
image recognition
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
5.2
Papers:
9.7K
Citations:
2.2W
Organization
No organization information available