| Presentation | 2022-03-10 Improving accuracy by optimizing activation function for convolutional neural network using homomorphic encryption Kohei Yagyu, Ren Takeuchi, Vo Ngoc Khoi Nguyen, Masakatsu Nishigaki, Tetsushi Ohki, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The development of secure neural network technology that performs prediction while encrypting data using homomorphic encryption is in progress. However, due to the unavailability of general activation functions, the classification accuracy of the secure neural network is low. In this paper, we investigate a method to generate activation functions that contribute to higher classification accuracy by pre-training using optimization. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Secure Neural Network / Convolutional Neural Network / Activation Function |
| Paper # | IT2021-92,ISEC2021-57,WBS2021-60,RCC2021-67 |
| Date of Issue | 2022-03-03 (IT, ISEC, WBS, RCC) |
| Conference Information | |
| Committee | IT / ISEC / RCC / WBS |
|---|---|
| Conference Date | 2022/3/10(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Joint Meeting of ISEC, IT, RCC, and WBS |
| Chair | Tadashi Wadayama(Nagoya Inst. of Tech.) / Tetsuya Izu(Fujitsu Labs.) / HUAN-BANG LI(NICT) / Masanori Hamamura(Kochi Univ. of Tech.) |
| Vice Chair | Tetsuya Kojima(Tokyo Kosen) / Noboru Kunihiro(Tsukuba Univ.) / Goichiro Hanaoka(AIST) / Shunichi Azuma(Nagoya Univ.) / Koji Ishii(Kagawa Univ.) / Takashi Shono(INTEL) / Masahiro Fujii(Utsunomiya Univ.) |
| Secretary | Tetsuya Kojima(Saitamai Univ.) / Noboru Kunihiro(Yamaguchi Univ.) / Goichiro Hanaoka(Fujitsu Labs.) / Shunichi Azuma(Ibaraki Univ.) / Koji Ishii(CRIEPI) / Takashi Shono(Osaka Univ.) / Masahiro Fujii(National Defence Academy) |
| Assistant | Masanori Hirotomo(Saga Univ.) / Takahiro Matsuda(AIST) / SHAN LIN(NICT) / Masaki Ogura(Osaka Univ.) / Masayuki Kinoshita(Chiba Univ. of Tech.) / Sun Ran(Ibaraki Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Information Theory / Technical Committee on Information Security / Technical Committee on Reliable Communication and Control / Technical Committee on Wideband System |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Improving accuracy by optimizing activation function for convolutional neural network using homomorphic encryption |
| Sub Title (in English) | |
| Keyword(1) | Secure Neural Network |
| Keyword(2) | Convolutional Neural Network |
| Keyword(3) | Activation Function |
| 1st Author's Name | Kohei Yagyu |
| 1st Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| 2nd Author's Name | Ren Takeuchi |
| 2nd Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| 3rd Author's Name | Vo Ngoc Khoi Nguyen |
| 3rd Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| 4th Author's Name | Masakatsu Nishigaki |
| 4th Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| 5th Author's Name | Tetsushi Ohki |
| 5th Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| Date | 2022-03-10 |
| Paper # | IT2021-92,ISEC2021-57,WBS2021-60,RCC2021-67 |
| Volume (vol) | vol.121 |
| Number (no) | IT-428,ISEC-429,WBS-430,RCC-431 |
| Page | pp.pp.58-65(IT), pp.58-65(ISEC), pp.58-65(WBS), pp.58-65(RCC), |
| #Pages | 8 |
| Date of Issue | 2022-03-03 (IT, ISEC, WBS, RCC) |