Boosting Adversarial Training With Mitigating Hard Sample Interference

已完结 10 由星发布于 2026/8/1 17:32:00

DOI:10.1109/TNNLS.2025.3648421
文献类型：期刊论文
补充材料：只需要正文
文献链接：https://doi.org/10.1109/TNNLS.2025.3648421

Institute of Electrical and Electronics Engineers (IEEE)

Abstract：Adversarial training (AT) has shown impressive advantages in maintaining accuracy and enhancing robustness against adversarial examples. However, most existing AT techniques jointly optimize

应助信息

7小时前

打酱油

ebc196dc8d9211f18f8234735aa3bc31.pdf 已采纳

7小时前

Book学术机器人

2026/8/1 17:32:01 未找到该文献，机器人已退出，请等待人工下载

2026/8/1 17:32:01 Book学术AI机器人收到请求，开始寻找文献

2026/8/1 17:32:01 已向机器人发送请求

7小时前

星发布求助

上一个求助:其它数据库Naturally Fractured Carbonate Reservoir Characterization: A Case Study of a Mature High-Pour Point Oil Field in Hungary 下一个求助:ElsevierMWCNTs-Beta-Cyclodextrin-reduced graphene oxide gel based electrochemical sensor for simultaneous detection of dopamine and uric acid in human sweat samples