Ajou University repository

Demo: EQuaTE: Efficient Quantum Train Engine Design and Demonstration for Dynamic Software Analysis
  • Park, Soohyun ;
  • Feng, Hao ;
  • Yun, Won Joon ;
  • Park, Chanyoung ;
  • Lee, Youn Kyu ;
  • Jung, Soyi ;
  • Kim, Joongheon
Citations

SCOPUS

0

Citation Export

Publication Year
2023-01-01
Journal
Proceedings - International Conference on Distributed Computing Systems
Publisher
Institute of Electrical and Electronics Engineers Inc.
Citation
Proceedings - International Conference on Distributed Computing Systems, Vol.2023-July, pp.1009-1012
Mesh Keyword
Dynamic softwaresDynamics analysisEngine designGradient varianceLocal minimumsMachine learning softwareProbabilisticsQuantum machinesQuantum neural networksSoftware analysis
All Science Classification Codes (ASJC)
SoftwareHardware and ArchitectureComputer Networks and Communications
Abstract
This paper proposes an efficient quantum train engine (EQuaTE), a novel tool for quantum machine learning software which plots gradient variances to check whether our quantum neural network (QNN) falls into local minima (called barren plateaus in QNN). EQuaTE can be realized via dynamic analysis of the undetermined probabilistic qubit states. Furthermore, the proposed EQuaTE is capable of HCI-based visual feedback such that software engineers can recognize barren plateaus via visualization, allowing the modification of QNN based on this information.
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/36951
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85174663769&origin=inward
DOI
https://doi.org/10.1109/icdcs57875.2023.00116
Type
Conference
Funding
V. CONCLUDINGREMARKS This paper proposes a novel QML software tool which visually plots gradient variances to check for the occurrence of barren plateaus via dynamic analysis, called EQuaTE. Furthermore, our EQuaTE is capable for visual feedback because it allows software engineers to modify QNNs based on the recognized barren plateaus using tensorboard. Acknowledgement. This work was supported by RF-Korea (2022R1A2C2004869). (Corresponding author: Joongheon Kim, joongheon@korea.ac.kr).
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Jung, Soyi Image
Jung, Soyi정소이
Department of Electrical and Computer Engineering
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.