| Presentation | 2022-05-27 Considerations for Identifying Machine Learning Models Naoto Kiribuchi, Ryohei Suzuki, Nami Ashizawa, Tetsushi Ohki, Hiroshi Mineno, Masakatsu Nishigaki, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | As a part of considerations for identifying machine learning models, we experimented whether we can identify multiple different models by their behaviors. We identify general software using something like a hash value provided by its producer who knows its behavior. However, machine learning models keep changing itself by continues learning, and the producer doesn't always know contents of the change. Thus, we considered to deal with model's continues changes and validate intended behaviors at the same time by checking its behaviors as the suitable identification method for machine learning models. In our experiments, we confirmed that we can distinguish multiple image recognition models using recognized results and its confidence for images with various blur intensities and brightness. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Machine Learning / Deep Learning / Neural Network / Identification / Authenticity / Image Recognition |
| Paper # | SC2022-2 |
| Date of Issue | 2022-05-20 (SC) |
| Conference Information | |
| Committee | SC |
|---|---|
| Conference Date | 2022/5/27(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | AI Service and Digital Transformation, and general topics |
| Chair | Shinji Kikuchi(NIMS) |
| Vice Chair | Yoji Yamato(NTT) / Kosaku Kimura(Fujitsu) |
| Secretary | Yoji Yamato(Kobe Univ.) / Kosaku Kimura(Tokyo Univ. of Tech.) |
| Assistant | Shin Tezuka(Hitachi) / Takao Nakaguchi(KCGI) |
| Paper Information | |
| Registration To | Technical Committee on Service Computing |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Considerations for Identifying Machine Learning Models |
| Sub Title (in English) | |
| Keyword(1) | Machine Learning |
| Keyword(2) | Deep Learning |
| Keyword(3) | Neural Network |
| Keyword(4) | Identification |
| Keyword(5) | Authenticity |
| Keyword(6) | Image Recognition |
| 1st Author's Name | Naoto Kiribuchi |
| 1st Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 2nd Author's Name | Ryohei Suzuki |
| 2nd Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 3rd Author's Name | Nami Ashizawa |
| 3rd Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 4th Author's Name | Tetsushi Ohki |
| 4th Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| 5th Author's Name | Hiroshi Mineno |
| 5th Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| 6th Author's Name | Masakatsu Nishigaki |
| 6th Author's Affiliation | Shizuoka University(Shizuoka Univ.) |
| Date | 2022-05-27 |
| Paper # | SC2022-2 |
| Volume (vol) | vol.122 |
| Number (no) | SC-50 |
| Page | pp.pp.7-12(SC), |
| #Pages | 6 |
| Date of Issue | 2022-05-20 (SC) |