| Presentation | 2022-03-10 Adversarial Training: A Survey Hiroki Adachi, Tsubasa Hirakawa, Takayoshi Yamashita, Hironobu Fujiyoshi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Adversarial training (AT) is a training method that aims to obtain a robust model for defencing the adversarial attack by using adversarial examples (AEs). Although AT improves the robustness of the model to AEs, it significantly decreases the classification accuracy to natural samples. To overcome this problem, researchers proposed methods that approached from several perspectives. In this paper, we survey AT and systematically summarize about research trends of AT. Furthermore, we evaluate and compare the classification accuracy with the exact experimental details for the typical methods. Moreover, we visualize the low dimensional feature space of the model applied to each method and evaluate the feature representation using some quantitative evaluation indices. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Deep learning / Adversarial examples / Adversarial training / Survey |
| Paper # | PRMU2021-73 |
| Date of Issue | 2022-03-03 (PRMU) |
| Conference Information | |
| Committee | PRMU / IPSJ-CVIM |
|---|---|
| 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) | Differentiable rendering |
| Chair | Seiichi Uchida(Kyushu Univ.) |
| Vice Chair | Masakazu Iwamura(Osaka Pref. Univ.) / Mitsuru Anpai(Denso IT Lab.) |
| Secretary | Masakazu Iwamura(NTT) / Mitsuru Anpai(Tottori Univ.) |
| Assistant | Kouta Yamaguchi(CyberAgent) / Yusuke Matsui(Univ. of Tokyo) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding / Special Interest Group on Computer Vision and Image Media |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Adversarial Training: A Survey |
| Sub Title (in English) | |
| Keyword(1) | Deep learning |
| Keyword(2) | Adversarial examples |
| Keyword(3) | Adversarial training |
| Keyword(4) | Survey |
| 1st Author's Name | Hiroki Adachi |
| 1st Author's Affiliation | Chubu University(Chubu Univ.) |
| 2nd Author's Name | Tsubasa Hirakawa |
| 2nd Author's Affiliation | Chubu University(Chubu Univ.) |
| 3rd Author's Name | Takayoshi Yamashita |
| 3rd Author's Affiliation | Chubu University(Chubu Univ.) |
| 4th Author's Name | Hironobu Fujiyoshi |
| 4th Author's Affiliation | Chubu University(Chubu Univ.) |
| Date | 2022-03-10 |
| Paper # | PRMU2021-73 |
| Volume (vol) | vol.121 |
| Number (no) | PRMU-427 |
| Page | pp.pp.78-90(PRMU), |
| #Pages | 13 |
| Date of Issue | 2022-03-03 (PRMU) |