Hot storiesContinuously updated
研究表明核机器原生对抗鲁棒性优于神经网络
1 reports1 reporting sourcesUpdated 5 hours ago
Start here
AI overview
一项针对图像与表格分类任务的最新实证评估,通过白盒与黑盒逃逸攻击系统检验了核机器(Kernel Machines)、随机特征分类器以及递归特征机器(Recursive Feature Machines)的鲁棒性。研究发现,核机器在原生状态下比神经网络更具对抗鲁棒性,在面临对抗扰动时性能退化更加平缓。此外,该研究还提出了针对核机器的近似对抗训练算法。
AI-generated from reports · Updated 5 hours ago
Latest developmentsNov 6 13:00
研究证实核机器原生状态下抗扰动能力强于神经网络,并提出近似对抗训练算法。Report timeline
Follow the reports to see different sides of the story.
Nov 6
- arXiv 机器学习核机器对抗鲁棒性全面分析
核机器(Kernel Machines)在原生状态下比神经网络更具对抗鲁棒性,面对对抗扰动时性能退化更为平缓。一项针对图像与表格分类任务的实证评估,通过白盒与黑盒逃逸攻击检验了核机器、随机特征分类器及递归特征机器(Recursive Feature Machines)的鲁棒性,并提出了针对核机器的近似对抗训练算法。
Story heat trend
Current heat 9·Comparable peak 10(Nov 6 14:00)·Comparable change, last 24 hours–
Trends compare only the same parties observed continuously, so the scope can be narrower than current heat counts. Move the pointer or tap the chart to see hourly heat; use the left and right arrow keys on a keyboard.