A 1W8R 20T SRAM Codebook for 20% Energy Reduction in Mixed-Precision Deep-Learning Inference Processor System

Author:

Ohara Ryotaro1,Masaya Kabuto2,Taichi Masakazu1,Fukunaga Atsushi2,Yasuda Yuto1,Hamabe Riku1,Izumi Shintaro2,Kawaguchi Hiroshi2

Affiliation:

1. Kobe University,Graduate School of System Informatics,Kobe,Japan

2. Kobe University,Graduate School of Science, Technology and Innovation,Kobe,Japan

Publisher

IEEE

Reference13 articles.

1. A 1.15-TOPS 6.57-TOPS/W Neural Network Processor for Multi-Scale Object Detection With Reduced Convolutional Operations

2. EIE;Han;ACM SIGARCH Computer Architecture News,2016

3. Greedy search algorithm for partial quantization of convolutional neural networks inspired by submodular optimization

4. PyTorch: An Imperative Style, High-Performance Deep Learning Library;Paszke,2019

5. Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding;Han,2015

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Multi-Ported GC-eDRAM Bitcell with Dynamic Port Configuration and Refresh Mechanism;Journal of Low Power Electronics and Applications;2024-01-04

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3