Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme

Geonwoo Jeon; Mi Yeon Hong; Joon Soo Yoo; Ji Won Yoon; Baekkyung Song

doi:10.1109/ICTC58733.2023.10392349

Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme

Geonwoo Jeon, Mi Yeon Hong, Joon Soo Yoo, Ji Won Yoon, Baekkyung Song

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In the context of the proliferating AI landscape, driven by entities like ChatGPT, the demand for extensive data utilization for training has surged, raising concerns about unauthorized data aggregation and privacy breaches. Paradoxically, this surge in data consumption has raised a critical concern - the breach of privacy stemming from unauthorized data aggregation. Sensitive data categories, including credit card details, medical records, and geographical locations, are particularly vulnerable to misuse. Homomorphic Encryption (HE), a post-quantum attack-resistant cryptographic technique, addresses this concern by enabling secure computations on encrypted data. However, HE's potential is hindered by limitations in evaluation speed, particularly evident in high-dimensional data analysis. This paper introduces contributions, including efficient inverse matrix computation, tailored eigenvector extraction via the power method for the TFHE scheme, and eigenvalue calculation using the Rayleigh quotient within TFHE. The feasibility of applying LDA in the encrypted domain is demonstrated using Fast Fully Homomorphic Encryption over the Torus (TFHE) scheme.

Original language	English
Title of host publication	ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence
Subtitle of host publication	Exploring the Frontiers of ICT Innovation
Publisher	IEEE Computer Society
Pages	1462-1467
Number of pages	6
ISBN (Electronic)	9798350313277
DOIs	https://doi.org/10.1109/ICTC58733.2023.10392349
Publication status	Published - 2023
Event	14th International Conference on Information and Communication Technology Convergence, ICTC 2023 - Jeju Island, Korea, Republic of Duration: 2023 Oct 11 → 2023 Oct 13

Publication series

Name	International Conference on ICT Convergence
ISSN (Print)	2162-1233
ISSN (Electronic)	2162-1241

Conference

Conference	14th International Conference on Information and Communication Technology Convergence, ICTC 2023
Country/Territory	Korea, Republic of
City	Jeju Island
Period	23/10/11 → 23/10/13

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

Homomorphic Encryption
Linear Discriminant Analysis
Newton's method
Power method
TFHE

ASJC Scopus subject areas

Information Systems
Computer Networks and Communications

Access to Document

10.1109/ICTC58733.2023.10392349

Cite this

Jeon, G., Hong, M. Y., Soo Yoo, J., Yoon, J. W., & Song, B. (2023). Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme. In ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence: Exploring the Frontiers of ICT Innovation (pp. 1462-1467). (International Conference on ICT Convergence). IEEE Computer Society. https://doi.org/10.1109/ICTC58733.2023.10392349

Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme. / Jeon, Geonwoo; Hong, Mi Yeon; Soo Yoo, Joon et al.
ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence: Exploring the Frontiers of ICT Innovation. IEEE Computer Society, 2023. p. 1462-1467 (International Conference on ICT Convergence).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Jeon, G, Hong, MY, Soo Yoo, J, Yoon, JW & Song, B 2023, Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme. in ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence: Exploring the Frontiers of ICT Innovation. International Conference on ICT Convergence, IEEE Computer Society, pp. 1462-1467, 14th International Conference on Information and Communication Technology Convergence, ICTC 2023, Jeju Island, Korea, Republic of, 23/10/11. https://doi.org/10.1109/ICTC58733.2023.10392349

Jeon G, Hong MY, Soo Yoo J, Yoon JW, Song B. Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme. In ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence: Exploring the Frontiers of ICT Innovation. IEEE Computer Society. 2023. p. 1462-1467. (International Conference on ICT Convergence). doi: 10.1109/ICTC58733.2023.10392349

Jeon, Geonwoo ; Hong, Mi Yeon ; Soo Yoo, Joon et al. / Secure Dimensionality Reduction : Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme. ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence: Exploring the Frontiers of ICT Innovation. IEEE Computer Society, 2023. pp. 1462-1467 (International Conference on ICT Convergence).

@inproceedings{bc3ce00afe704a33bf3ca702f6ad5b0a,

title = "Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme",

abstract = "In the context of the proliferating AI landscape, driven by entities like ChatGPT, the demand for extensive data utilization for training has surged, raising concerns about unauthorized data aggregation and privacy breaches. Paradoxically, this surge in data consumption has raised a critical concern - the breach of privacy stemming from unauthorized data aggregation. Sensitive data categories, including credit card details, medical records, and geographical locations, are particularly vulnerable to misuse. Homomorphic Encryption (HE), a post-quantum attack-resistant cryptographic technique, addresses this concern by enabling secure computations on encrypted data. However, HE's potential is hindered by limitations in evaluation speed, particularly evident in high-dimensional data analysis. This paper introduces contributions, including efficient inverse matrix computation, tailored eigenvector extraction via the power method for the TFHE scheme, and eigenvalue calculation using the Rayleigh quotient within TFHE. The feasibility of applying LDA in the encrypted domain is demonstrated using Fast Fully Homomorphic Encryption over the Torus (TFHE) scheme.",

keywords = "Homomorphic Encryption, Linear Discriminant Analysis, Newton's method, Power method, TFHE",

author = "Geonwoo Jeon and Hong, {Mi Yeon} and {Soo Yoo}, Joon and Yoon, {Ji Won} and Baekkyung Song",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 14th International Conference on Information and Communication Technology Convergence, ICTC 2023 ; Conference date: 11-10-2023 Through 13-10-2023",

year = "2023",

doi = "10.1109/ICTC58733.2023.10392349",

language = "English",

series = "International Conference on ICT Convergence",

publisher = "IEEE Computer Society",

pages = "1462--1467",

booktitle = "ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence",

}

TY - GEN

T1 - Secure Dimensionality Reduction

T2 - 14th International Conference on Information and Communication Technology Convergence, ICTC 2023

AU - Jeon, Geonwoo

AU - Hong, Mi Yeon

AU - Soo Yoo, Joon

AU - Yoon, Ji Won

AU - Song, Baekkyung

PY - 2023

Y1 - 2023

N2 - In the context of the proliferating AI landscape, driven by entities like ChatGPT, the demand for extensive data utilization for training has surged, raising concerns about unauthorized data aggregation and privacy breaches. Paradoxically, this surge in data consumption has raised a critical concern - the breach of privacy stemming from unauthorized data aggregation. Sensitive data categories, including credit card details, medical records, and geographical locations, are particularly vulnerable to misuse. Homomorphic Encryption (HE), a post-quantum attack-resistant cryptographic technique, addresses this concern by enabling secure computations on encrypted data. However, HE's potential is hindered by limitations in evaluation speed, particularly evident in high-dimensional data analysis. This paper introduces contributions, including efficient inverse matrix computation, tailored eigenvector extraction via the power method for the TFHE scheme, and eigenvalue calculation using the Rayleigh quotient within TFHE. The feasibility of applying LDA in the encrypted domain is demonstrated using Fast Fully Homomorphic Encryption over the Torus (TFHE) scheme.

AB - In the context of the proliferating AI landscape, driven by entities like ChatGPT, the demand for extensive data utilization for training has surged, raising concerns about unauthorized data aggregation and privacy breaches. Paradoxically, this surge in data consumption has raised a critical concern - the breach of privacy stemming from unauthorized data aggregation. Sensitive data categories, including credit card details, medical records, and geographical locations, are particularly vulnerable to misuse. Homomorphic Encryption (HE), a post-quantum attack-resistant cryptographic technique, addresses this concern by enabling secure computations on encrypted data. However, HE's potential is hindered by limitations in evaluation speed, particularly evident in high-dimensional data analysis. This paper introduces contributions, including efficient inverse matrix computation, tailored eigenvector extraction via the power method for the TFHE scheme, and eigenvalue calculation using the Rayleigh quotient within TFHE. The feasibility of applying LDA in the encrypted domain is demonstrated using Fast Fully Homomorphic Encryption over the Torus (TFHE) scheme.

KW - Homomorphic Encryption

KW - Linear Discriminant Analysis

KW - Newton's method

KW - Power method

KW - TFHE

UR - http://www.scopus.com/inward/record.url?scp=85184615850&partnerID=8YFLogxK

U2 - 10.1109/ICTC58733.2023.10392349

DO - 10.1109/ICTC58733.2023.10392349

M3 - Conference contribution

AN - SCOPUS:85184615850

T3 - International Conference on ICT Convergence

SP - 1462

EP - 1467

BT - ICTC 2023 - 14th International Conference on Information and Communication Technology Convergence

PB - IEEE Computer Society

Y2 - 11 October 2023 through 13 October 2023

ER -

Secure Dimensionality Reduction: Applying Linear Discriminant Analysis over the TFHE Homomorphic Encryption Scheme

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this