AIMC Modeling and Parameter Tuning for Layer-Wise Optimal Operating Point in DNN Inference

Author:

Dadras Iman1ORCID,Sarda Giuseppe M.2ORCID,Laubeuf Nathan2,Bhattacharjee Debjyoti3ORCID,Mallik Arindam3

Affiliation:

1. Intelligent Materials and Systems Laboratory (IMS Laboratory), Institute of Technology, University of Tartu, Tartu, Estonia

2. ESAT-MICAS, KU Leuven, Leuven, Belgium

3. imec, Leuven, Belgium

Funder

European Research Council

European Union’s Horizon 2020 Research and Innovation Program

Flanders AI Research Program, Katholieke Universiteit (KU) Leuven, Estonian Research Council

Estonian Centre of Excellence in ICT Research and Doctoral School of the European Institute of Innovation and Technology (EIT) Manufacturing, funded by the European Union

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

General Engineering,General Materials Science,General Computer Science,Electrical and Electronic Engineering

Reference43 articles.

1. Discovering low-precision networks close to full-precision networks for efficient embedded inference;mckinstry;arXiv 1809 04191,2018

2. Accelerating CNN Inference on ASICs: A Survey

3. Adaptive block floating-point for analog deep learning hardware;basumallik;arXiv 2205 06287,2022

4. An Overview of Processing-in-Memory Circuits for Artificial Intelligence and Machine Learning

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