Automated Neural Network Accelerator Generation Framework for Multiple Neural Network Applications
- Authors
- Lee, Inho; Hong, Seongmin; Ryu, Giha; Park, Yongjun
- Issue Date
- Oct-2019
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- accelerator; FPGA; neural network
- Citation
- IEEE Region 10 Annual International Conference, Proceedings/TENCON, v.2018-October, pp.2287 - 2290
- Indexed
- SCOPUS
- Journal Title
- IEEE Region 10 Annual International Conference, Proceedings/TENCON
- Volume
- 2018-October
- Start Page
- 2287
- End Page
- 2290
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/147092
- DOI
- 10.1109/TENCON.2018.8650190
- ISSN
- 2159-3442
- Abstract
- Neural networks are widely used in various applications, but general neural network accelerators support only one application at a time. Therefore, information for each application, such as synaptic weights and bias data, must be loaded quickly to use multiple neural network applications. Field-programmable gate array (FPGA)-based implementation has huge performance overhead owing to low data transmission bandwidth. In order to solve this problem, this paper presents an automated FPGA-based multi-neural network accelerator generation framework that can quickly support several applications by storing neural network application data in an on-chip memory inside the FPGA. To do this, we first design a shared custom hardware accelerator that can support rapid changes in multiple target neural network applications. Then, we introduce an automated multi-neural network accelerator generation framework that performs training, weight quantization, and neural accelerator synthesis.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 컴퓨터소프트웨어학부 > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/147092)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.