Presentation 2022-03-08
RUT Computation of Gaussian Process Regression for Encrypting Input and Output Signals
Takayuki Nakachi, Yitu Wang,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) n this paper, we propose Gaussian Process Regression (GPR) for encrypted data generated based on random unitary transformation (RUT) and evaluate its performance. In recent years, computational forms that utilize computational resources provided by providers using edge and cloud services have rapidly become widespread. However, there are concerns about problems such as data fraud, leakage, and privacy invasion due to lack of reliability of providers and accidents. This paper examines the secure Gaussian process regression method that takes privacy protection into consideration. It is possible to solve the Gaussian process regression problem by encrypting not only the input signal but also the output signal. It is shown that the prediction performance does not deteriorate as compared with the case without encryption. Finally, as an application example of the secure Gaussian process regression, the disease progression is predicted using clinical data of diabetes, and the effectiveness of the proposed method is verified by simulation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Machine Learning / Gaussian Process / Random Unitary Transform / Secure Computation / Edge/Cloud Computing
Paper # EMM2021-115
Date of Issue 2022-02-28 (EMM)

Conference Information
Committee EMM
Conference Date 2022/3/7(2days)
Place (in Japanese) (See Japanese page)
Place (in English) (Primary: Online, Secondary: On-site)
Topics (in Japanese) (See Japanese page)
Topics (in English) Image and Sound Quality, Metrics for Perception and Recognition, Human Auditory and Visual System, etc.
Chair Ryoichi Nishimura(NICT)
Vice Chair Masaaki Fujiyoshi(Tokyo Metropolitan Univ.) / Masatsugu Ichino(Univ. of Electro-Comm.)
Secretary Masaaki Fujiyoshi(Utsunomiya Univ.) / Masatsugu Ichino(NICT)
Assistant Shoko Imaizumi(Chiba Univ.) / Youichi Takashima(Kaishi Professional Univ.)

Paper Information
Registration To Technical Committee on Enriched MultiMedia
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) RUT Computation of Gaussian Process Regression for Encrypting Input and Output Signals
Sub Title (in English)
Keyword(1) Machine Learning
Keyword(2) Gaussian Process
Keyword(3) Random Unitary Transform
Keyword(4) Secure Computation
Keyword(5) Edge/Cloud Computing
1st Author's Name Takayuki Nakachi
1st Author's Affiliation University of the Ryukyus(Univ. of the Ryukyus)
2nd Author's Name Yitu Wang
2nd Author's Affiliation Nippon Telegraph and Telephone Corporation(NTT)
Date 2022-03-08
Paper # EMM2021-115
Volume (vol) vol.121
Number (no) EMM-417
Page pp.pp.124-129(EMM),
#Pages 6
Date of Issue 2022-02-28 (EMM)