Wing W Y Ng1, Qin Zhang1, Cankun Zhong2

  • 1Guangdong Provincial Key Laboratory of Computational Intelligence and Cyberspace Information, School of Computer Science & Engineer, South China University of Technology, Guangzhou, 510006, China.

概括

这项研究引入了一种新的域泛化方法,使用混合域注意力来关注重要的视觉特征. 该方法通过降低对输入扰动的灵敏度来提高模型的稳定性和概括性.