Title:
SOFTMAX FUNCTION SECRET CALCULATION SYSTEM, SOFTMAX FUNCTION SECRET CALCULATION DEVICE, SOFTMAX FUNCTION SECRET CALCULATION METHOD, NEURAL NETWORK SECRET CALCULATION SYSTEM, NEURAL NETWORK SECRET LEARNING SYSTEM, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2021/029024
Kind Code:
A1
Abstract:
Provided is technology capable of performing high-speed and high-accuracy softmax function secret calculation. A softmax function secret calculation system for calculating a share ([[softmax(u1)]], …, [[softmax(uJ)]]) from a share ([[u1]], …, [[uJ]]) comprises: a subtraction means for calculating a share ([[u1-u1]], [[u2-u1]], …, [[uJ-uJ]]); a first batch mapping secret calculation means for calculating ([[exp(u1-u1)]], [[exp(u2-u1)]], …, [[exp(uJ-uJ)]]); an adding means for calculating a share ([[Σj=1
Jexp(uj-u1)]], …, [[Σj=1
Jexp(uj-uJ)]]); and a second batch mapping secret calculation means for calculating a share ([[softmax(u1)]], …, [[softmax(uJ)]]).
More Like This:
Inventors:
MISHINA IBUKI (JP)
IKARASHI DAI (JP)
HAMADA KOKI (JP)
IKARASHI DAI (JP)
HAMADA KOKI (JP)
Application Number:
PCT/JP2019/031916
Publication Date:
February 18, 2021
Filing Date:
August 14, 2019
Export Citation:
Assignee:
NIPPON TELEGRAPH & TELEPHONE (JP)
International Classes:
G09C1/00; G06F21/60
Other References:
TANAKA, SATOSHI ET AL.: "Cost Evaluation of Neural Network using Secret-sharing- based Secure Computation", IEICE TECHNICAL REPORT, vol. 116, no. 65, 19 May 2016 (2016-05-19), pages 119 - 126, XP009523659
SATSUYA OHATA: "Round-Efficient Secure Two-Party Computation and Its Application to Privacy-Preserving Convolutional Neural Networks", PROCEEDINGS OF COMPUTER SECURITY SYMPOSIUM 2018, vol. 2018, no. 2, 15 October 2018 (2018-10-15) - 25 October 2018 (2018-10-25), Japan , pages 615 - 622, XP009533503, ISSN: 1882-0840
MISHINA, IBUKI ET AL.: "Designs and Implementations of Efficient and Accurate Secret Logistic Regression", PROCEEDINGS OF COMPUTER SECURITY SYMPOSIUM 2018, vol. 2018, no. 2, 15 October 2018 (2018-10-15), pages 1229 - 1236, XP009522747
MISHINA, IBUKI ET AL.: "Can the Logistic Regression in Secure Computation Really be Used?", PREPRINTS OF THE 2019 SYMPOSIUM ON CRYPTOGRAPHY AND INFORMATION SECURITY, 31 December 2018 (2018-12-31) - 25 January 2019 (2019-01-25), JP, pages 1 - 8, XP009530138
PAYMAN MOHASSELYUPENG ZHANG: "SecureML: A System for Scalable Privacy-Preserving Machine Learning", 2017 IEEE SYMPOSIUM ON SECURITY AND PRIVACY, 2017, pages 19 - 38, XP055554322, DOI: 10.1109/SP.2017.12
SAMEER WAGHDIVYA GUPTANISHANTH CHANDRAN: "SecureNN: 3-Party Secure Computation for Neural Network Training", PROCEEDINGS ON PRIVACY ENHANCING TECHNOLOGIES, vol. 2019, no. 3, 2019, pages 26 - 49
SHAMIR, A.: "How to share a secret", COMMUNICATIONS OF THE ACM, vol. 22, no. 11, 1979, pages 612 - 613, XP000565227, DOI: 10.1145/359168.359176
BEN-OR, M.GOLDWASSER, S.WIGDERSON, A.: "Proceedings of the twentieth annual ACM symposium on Theory of computing", 1988, ACM, article "Completeness theorems for non-cryptographic fault-tolerant distributed computation", pages: 1 - 10
GENNARO, R.RABIN, M. O.RABIN, T.: "Proceedings of the seventeenth annual ACM symposium on Principles of distributed computing", 1998, ACM, article "Simplied VSS and fast-track multiparty computations with applications to threshold cryptography", pages: 101 - 111
KOKI HAMADADAI IKARASHIKOJI CHIDA: "A Batch Mapping Algorithm for Secure Function Evaluation", IEICE TRANSACTIONS A, vol. J96, no. 4, 2013, pages 157 - 165, XP008184067
See also references of EP 4016506A4
SATSUYA OHATA: "Round-Efficient Secure Two-Party Computation and Its Application to Privacy-Preserving Convolutional Neural Networks", PROCEEDINGS OF COMPUTER SECURITY SYMPOSIUM 2018, vol. 2018, no. 2, 15 October 2018 (2018-10-15) - 25 October 2018 (2018-10-25), Japan , pages 615 - 622, XP009533503, ISSN: 1882-0840
MISHINA, IBUKI ET AL.: "Designs and Implementations of Efficient and Accurate Secret Logistic Regression", PROCEEDINGS OF COMPUTER SECURITY SYMPOSIUM 2018, vol. 2018, no. 2, 15 October 2018 (2018-10-15), pages 1229 - 1236, XP009522747
MISHINA, IBUKI ET AL.: "Can the Logistic Regression in Secure Computation Really be Used?", PREPRINTS OF THE 2019 SYMPOSIUM ON CRYPTOGRAPHY AND INFORMATION SECURITY, 31 December 2018 (2018-12-31) - 25 January 2019 (2019-01-25), JP, pages 1 - 8, XP009530138
PAYMAN MOHASSELYUPENG ZHANG: "SecureML: A System for Scalable Privacy-Preserving Machine Learning", 2017 IEEE SYMPOSIUM ON SECURITY AND PRIVACY, 2017, pages 19 - 38, XP055554322, DOI: 10.1109/SP.2017.12
SAMEER WAGHDIVYA GUPTANISHANTH CHANDRAN: "SecureNN: 3-Party Secure Computation for Neural Network Training", PROCEEDINGS ON PRIVACY ENHANCING TECHNOLOGIES, vol. 2019, no. 3, 2019, pages 26 - 49
SHAMIR, A.: "How to share a secret", COMMUNICATIONS OF THE ACM, vol. 22, no. 11, 1979, pages 612 - 613, XP000565227, DOI: 10.1145/359168.359176
BEN-OR, M.GOLDWASSER, S.WIGDERSON, A.: "Proceedings of the twentieth annual ACM symposium on Theory of computing", 1988, ACM, article "Completeness theorems for non-cryptographic fault-tolerant distributed computation", pages: 1 - 10
GENNARO, R.RABIN, M. O.RABIN, T.: "Proceedings of the seventeenth annual ACM symposium on Principles of distributed computing", 1998, ACM, article "Simplied VSS and fast-track multiparty computations with applications to threshold cryptography", pages: 101 - 111
KOKI HAMADADAI IKARASHIKOJI CHIDA: "A Batch Mapping Algorithm for Secure Function Evaluation", IEICE TRANSACTIONS A, vol. J96, no. 4, 2013, pages 157 - 165, XP008184067
See also references of EP 4016506A4
Attorney, Agent or Firm:
NAKAO, Naoki et al. (JP)
Download PDF:
Previous Patent: USER DEVICE, BASE STATION DEVICE, AND COMMUNICATION METHOD
Next Patent: DEFECT INSPECTION DEVICE AND DEFECT INSPECTION METHOD
Next Patent: DEFECT INSPECTION DEVICE AND DEFECT INSPECTION METHOD