Deep Learning on Cryptocurrency-Challenges Processing GPU and CPU Architectures

Alex Trawinski, Hayden Wimmer, Weitian Tong

Research output: Contribution to book or proceedingConference articlepeer-review

Abstract

Deep learning has made vast advances over the recent decade due to hardware, software, and architecture improvements. Applying deep learning still has some challenges to overcome, especially when processing on GPU or in parallel. This work seeks to apply RNN and LSTM over a variety of architectures to demonstrate the challenges and techniques encountered when applying deep learning. We apply RNN and LSTM over GPU and CPU across PC, high performance computing cluster, and cloud and present our challenges and results.

Original languageEnglish
Title of host publication2021 IEEE 12th Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2021
EditorsSatyajit Chakrabarti, Rajashree Paul
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages371-377
Number of pages7
ISBN (Electronic)9781665400664
DOIs
StatePublished - 2021
Event12th IEEE Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2021 - Vancouver, Canada
Duration: Oct 27 2021Oct 30 2021

Publication series

Name2021 IEEE 12th Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2021

Conference

Conference12th IEEE Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2021
Country/TerritoryCanada
CityVancouver
Period10/27/2110/30/21

Scopus Subject Areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Hardware and Architecture
  • Information Systems
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality

Keywords

  • Bitcoin
  • Cryptocurrency
  • Deep Learning
  • GPU
  • LSTM
  • RNN

Fingerprint

Dive into the research topics of 'Deep Learning on Cryptocurrency-Challenges Processing GPU and CPU Architectures'. Together they form a unique fingerprint.

Cite this