Modifying the one-hot encoding technique can enhance the adversarial robustness of the visual model for symbol recognition

Yi Sun, Jun Zheng, Hanyu Zhao, Huipeng Zhou, Jiaxing Li, Fan Li, Zehui Xiong, Jun Liu, Yuanzhang Li*

*此作品的通讯作者

科研成果: 期刊稿件 › 文章 › 同行评审

3 引用 （Scopus）

指纹

探究 'Modifying the one-hot encoding technique can enhance the adversarial robustness of the visual model for symbol recognition' 的科研主题。它们共同构成独一无二的指纹。

Engineering

Computer Science