Modifying the one-hot encoding technique can enhance the adversarial robustness of the visual model for symbol recognition
Autor: | Sun, Yi, Zheng, Jun, Zhao, Hanyu, Zhou, Huipeng, Li, Jiaxing, Li, Fan, Xiong, Zehui, Liu, Jun, Li, Yuanzhang |
---|---|
Zdroj: | In Expert Systems With Applications 15 September 2024 250 |
Databáze: | ScienceDirect |
Externí odkaz: |