通过滑动窗口对抗训练，逐步缩小域间差距，提升渐进域适应性能。
SWAT: Sliding Window Adversarial Training for Gradual Domain Adaptation
- 构建对抗流连接源域与目标域特征空间，形成渐进适配路径
- 滑动窗口逐段缩小相邻中间域的差异，最终使目标域域偏移显著降低
- 在6个基准上验证有效，尤其在旋转MNIST上提升6.1%
域偏移是影响机器学习性能的关键问题。无监督域自适应（UDA）虽能缓解该问题，但在域偏移剧烈时表现不佳。渐进域适应（GDA）通过多个中间域，以温和方式从源域逐步适应至目标域，缓解此问题。本文提出滑动窗口对抗训练（SWAT）用于GDA：首先构建对抗流，连接源域与目标域的特征空间；随后设计滑动窗口机制，沿对抗流移动，逐步缩小相邻中间域间的差距。当窗口抵达流末端（即目标域）时，域偏移被显式减小。在六个GDA基准上的大量实验表明，SWAT显著有效，尤其在旋转MNIST上比之前方法提升6.1%，在CIFAR-100C上提升4.1%。
原文摘要 · Abstract (English)
Domain shifts are critical issues that harm the performance of machine learning. Unsupervised Domain Adaptation (UDA) mitigates this issue but suffers when the domain shifts are steep and drastic. Gradual Domain Adaptation (GDA) alleviates this problem in a mild way by gradually adapting from the source to the target domain using multiple intermediate domains. In this paper, we propose Sliding Window Adversarial Training (SWAT) for GDA. SWAT first formulates adversarial streams to connect the feature spaces of the source and target domains. Then, a sliding window paradigm is designed that moves along the adversarial stream to gradually narrow the small gap between adjacent intermediate domains. When the window moves to the end of the stream, i.e., the target domain, the domain shift is explicitly reduced. Extensive experiments on six GDA benchmarks demonstrate the significant effectiveness of SWAT, especially 6.1% improvement on Rotated MNIST and 4.1% advantage on CIFAR-100C over the previous methods.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。