TrainWare: A memory optimized weight update architecture for on-device convolutional neural network training

Seungkyu Choi, Jaehyeong Sim, Myeonggu Kang, Lee Sup Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations

Abstract

Training convolutional neural network on device has become essential where it allows applications to consider user's individual environment. Meanwhile, the weight update operation from the training process is the primary factor of high energy consumption due to its substantial memory accesses. We propose a dedicated weight update architecture with two key features: (1) a specialized local buffer for the DRAM access deduction (2) a novel dataflow and its suitable processing element array structure for weight gradient computation to optimize the energy consumed by internal memories. Our scheme achieves 14.3%-30.2% total energy reduction by drastically eliminating the memory accesses.

Original languageEnglish
Title of host publicationISLPED 2018 - Proceedings of the 2018 International Symposium on Low Power Electronics and Design
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Print)9781450357043
DOIs
StatePublished - 23 Jul 2018
Event23rd IEEE/ACM International Symposium on Low Power Electronics and Design, ISLPED 2018 - Bellevue, United States
Duration: 23 Jul 201825 Jul 2018

Publication series

NameProceedings of the International Symposium on Low Power Electronics and Design
ISSN (Print)1533-4678

Conference

Conference23rd IEEE/ACM International Symposium on Low Power Electronics and Design, ISLPED 2018
Country/TerritoryUnited States
CityBellevue
Period23/07/1825/07/18

Keywords

  • CNN
  • Dataflow
  • On-device
  • Training architecture
  • Weight update

Fingerprint

Dive into the research topics of 'TrainWare: A memory optimized weight update architecture for on-device convolutional neural network training'. Together they form a unique fingerprint.

Cite this