A FPGA-based neural accelerator for small IoT devices

Hong, Seongmin; Park, Yongjun

doi:10.1109/ISOCC.2017.8368903

상세 보기

A FPGA-based neural accelerator for small IoT devices

Hong, Seongmin;
Park, Yongjun

Citations

SCOPUS

9

초록

Neural network has been widely used for various applications. While most of previous approaches tried to use large neural networks such as convolutional neural network (CNN) and deep neural network (DNN), these heavy models are hardly adapted to IoT(internet of things) platforms due to their limited resources. This work proposes a compact neural network accelerator for IoT devices. Our design shows 11.95 GOP/s total throughput and 413.99mW power consumption with 98.04% accuracy.

키워드

Accelerator; FPGA; Neural networks; Deep neural networks; Field programmable gate arrays (FPGA); Neural networks; Particle accelerators; Convolutional Neural Networks (CNN); Iot devices; Internet of things

제목: A FPGA-based neural accelerator for small IoT devices

저자: Hong, Seongmin; Park, Yongjun

DOI: 10.1109/ISOCC.2017.8368903

발행일: 2018-05

유형: Conference Paper

저널명: Proceedings - International SoC Design Conference 2017, ISOCC 2017

페이지: 294 ~ 295