Fast batch gradient descent in quantum neural networks

Joo Yong Shim; Joongheon Kim

doi:10.1049/ell2.70162

Fast batch gradient descent in quantum neural networks

Joo Yong Shim
, Joongheon Kim^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

A novel batch gradient descent algorithm for parameterized quantum circuits that significantly reduces the time complexity in terms of batch size for training quantum neural networks is proposed. Batch data constructed to quantum random access memory (qRAM) structure is mapped to one circuit that estimates average loss. As the number of circuits decreases, the range to which quantum amplitude estimation can be applied increases, speeding up with a quadratic scale in batch size.

Original language	English
Article number	e70162
Journal	Electronics Letters
Volume	61
Issue number	1
DOIs	https://doi.org/10.1049/ell2.70162
Publication status	Published - 2025 Jan 1

Bibliographical note

Publisher Copyright:
© 2025 The Author(s). Electronics Letters published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

Keywords

computational complexity
gradient methods
learning (artificial intelligence)
quantum computing

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1049/ell2.70162

Cite this

@article{c216a2399f3047768d846d39e13e59f5,

title = "Fast batch gradient descent in quantum neural networks",

abstract = "A novel batch gradient descent algorithm for parameterized quantum circuits that significantly reduces the time complexity in terms of batch size for training quantum neural networks is proposed. Batch data constructed to quantum random access memory (qRAM) structure is mapped to one circuit that estimates average loss. As the number of circuits decreases, the range to which quantum amplitude estimation can be applied increases, speeding up with a quadratic scale in batch size.",

keywords = "computational complexity, gradient methods, learning (artificial intelligence), quantum computing",

author = "Shim, \{Joo Yong\} and Joongheon Kim",

note = "Publisher Copyright: {\textcopyright} 2025 The Author(s). Electronics Letters published by John Wiley \& Sons Ltd on behalf of The Institution of Engineering and Technology.",

year = "2025",

month = jan,

day = "1",

doi = "10.1049/ell2.70162",

language = "English",

volume = "61",

journal = "Electronics Letters",

issn = "0013-5194",

publisher = "John Wiley \& Sons Inc.",

number = "1",

}

TY - JOUR

T1 - Fast batch gradient descent in quantum neural networks

AU - Shim, Joo Yong

AU - Kim, Joongheon

PY - 2025/1/1

Y1 - 2025/1/1

N2 - A novel batch gradient descent algorithm for parameterized quantum circuits that significantly reduces the time complexity in terms of batch size for training quantum neural networks is proposed. Batch data constructed to quantum random access memory (qRAM) structure is mapped to one circuit that estimates average loss. As the number of circuits decreases, the range to which quantum amplitude estimation can be applied increases, speeding up with a quadratic scale in batch size.

AB - A novel batch gradient descent algorithm for parameterized quantum circuits that significantly reduces the time complexity in terms of batch size for training quantum neural networks is proposed. Batch data constructed to quantum random access memory (qRAM) structure is mapped to one circuit that estimates average loss. As the number of circuits decreases, the range to which quantum amplitude estimation can be applied increases, speeding up with a quadratic scale in batch size.

KW - computational complexity

KW - gradient methods

KW - learning (artificial intelligence)

KW - quantum computing

UR - https://www.scopus.com/pages/publications/85216991595

U2 - 10.1049/ell2.70162

DO - 10.1049/ell2.70162

M3 - Article

AN - SCOPUS:85216991595

SN - 0013-5194

VL - 61

JO - Electronics Letters

JF - Electronics Letters

IS - 1

M1 - e70162

ER -

Fast batch gradient descent in quantum neural networks

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this