WEKO3
アイテム
Eliminating Adversarial Perturbations Using Image-to-Image Translation Method
http://hdl.handle.net/10228/0002000805
http://hdl.handle.net/10228/000200080587e7b5de-e7e2-4e3c-950c-1288c42ce726
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
10435531.pdf (10.2 MB)
|
|
| Item type | 学術雑誌論文 = Journal Article(1) | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2024-06-19 | |||||||||||||||||
| 資源タイプ | ||||||||||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||||||||||||||
| 資源タイプ | journal article | |||||||||||||||||
| タイトル | ||||||||||||||||||
| タイトル | Eliminating Adversarial Perturbations Using Image-to-Image Translation Method | |||||||||||||||||
| 言語 | en | |||||||||||||||||
| 言語 | ||||||||||||||||||
| 言語 | eng | |||||||||||||||||
| 著者 |
張, 海波
× 張, 海波
WEKO
35483
× Yao, Zhihua
× Sakurai, Kouichi
|
|||||||||||||||||
| 抄録 | ||||||||||||||||||
| 内容記述タイプ | Abstract | |||||||||||||||||
| 内容記述 | Convolutional neural networks are widely used for image recognition tasks, but they are vulnerable to adversarial attacks that can cause the model to misclassify an image. Such attacks pose a significant security risk in safety-critical applications like facial recognition and autonomous driving. Researchers have made progress in defending against adversarial attacks through two approaches: enhancing the neural networks themselves to be more robust and removing the perturbation added to the image through pre-processing. This paper is based upon a recent defense model that belongs to the latter approach, which utilizes image-to-image translation to regenerate images perturbed by adversarial attacks. We optimized the training process of their model and tested the model performance against more recent and strong attacks. The results show that the model is able to regenerate images attacked by the state-of-the-art attack, the AutoAttack, and restores the classification accuracy to a level over 83% to that of the original images. | |||||||||||||||||
| 言語 | en | |||||||||||||||||
| 備考 | ||||||||||||||||||
| 内容記述タイプ | Other | |||||||||||||||||
| 内容記述 | Applied Cryptography and Network Security Workshops (ACNS 2023) ,June 19–22, 2023, Kyoto, Japan | |||||||||||||||||
| 言語 | en | |||||||||||||||||
| 書誌情報 |
en : Lecture Notes in Computer Science 巻 13907, p. 601-620, 発行日 2023-10-04 |
|||||||||||||||||
| 出版社 | ||||||||||||||||||
| 出版者 | Springer | |||||||||||||||||
| DOI | ||||||||||||||||||
| 識別子タイプ | DOI | |||||||||||||||||
| 関連識別子 | https://doi.org/10.1007/978-3-031-41181-6_32 | |||||||||||||||||
| ISSN | ||||||||||||||||||
| 収録物識別子タイプ | PISSN | |||||||||||||||||
| 収録物識別子 | 0302-9743 | |||||||||||||||||
| ISSN | ||||||||||||||||||
| 収録物識別子タイプ | EISSN | |||||||||||||||||
| 収録物識別子 | 1611-3349 | |||||||||||||||||
| 著作権関連情報 | ||||||||||||||||||
| 権利情報 | Copyright (c) 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG. This is a post-peer-review, pre-copyedit version of an article published in Lecture Notes in Computer Science. The final authenticated version is available online at: https://doi.org/10.1007/978-3-031-41181-6_32. | |||||||||||||||||
| キーワード | ||||||||||||||||||
| 主題Scheme | Other | |||||||||||||||||
| 主題 | Adversarial attack | |||||||||||||||||
| キーワード | ||||||||||||||||||
| 主題Scheme | Other | |||||||||||||||||
| 主題 | Defense method | |||||||||||||||||
| キーワード | ||||||||||||||||||
| 主題Scheme | Other | |||||||||||||||||
| 主題 | Image-to-image translation | |||||||||||||||||
| 出版タイプ | ||||||||||||||||||
| 出版タイプ | AM | |||||||||||||||||
| 出版タイプResource | http://purl.org/coar/version/c_ab4af688f83e57aa | |||||||||||||||||
| 査読の有無 | ||||||||||||||||||
| 値 | yes | |||||||||||||||||
| 研究者情報 | ||||||||||||||||||
| URL | https://hyokadb02.jimu.kyutech.ac.jp/html/100001768_ja.html | |||||||||||||||||
| 論文ID（連携） | ||||||||||||||||||
| 値 | 10435531 | |||||||||||||||||
| 連携ID | ||||||||||||||||||
| 値 | 12348 | |||||||||||||||||
