论文标题
有条件域匹配和标签偏移的最佳传输
Optimal Transport for Conditional Domain Matching and Label Shift
论文作者
论文摘要
我们解决了在广义目标变化（联合类条件和标签偏移）设置下无监督域适应的问题。对于此框架，我们从理论上表明，为了良好的概括，有必要学习潜在表示，在该表示中，边际和阶级条件分布都跨域对齐。为此，我们提出了一个学习问题，该问题可最大程度地减少源域中的重要性加权损失，而加权边缘之间的瓦斯汀距离。为了进行适当的加权，我们通过将混合物估计和最佳匹配通过最佳传输来提供目标标签比例的估计器。该估计伴随着在轻度假设下正确性的理论保证。我们的实验结果表明，我们的方法的平均表现要比范围域适应性问题（包括\ emph {digits}，\ emph {visda}和\ emph {office}，平均表现更好。本文的代码可在\ url {https://github.com/arakotom/mars_domain_adaptation}获得。
We address the problem of unsupervised domain adaptation under the setting of generalized target shift (joint class-conditional and label shifts). For this framework, we theoretically show that, for good generalization, it is necessary to learn a latent representation in which both marginals and class-conditional distributions are aligned across domains. For this sake, we propose a learning problem that minimizes importance weighted loss in the source domain and a Wasserstein distance between weighted marginals. For a proper weighting, we provide an estimator of target label proportion by blending mixture estimation and optimal matching by optimal transport. This estimation comes with theoretical guarantees of correctness under mild assumptions. Our experimental results show that our method performs better on average than competitors across a range domain adaptation problems including \emph{digits},\emph{VisDA} and \emph{Office}. Code for this paper is available at \url{https://github.com/arakotom/mars_domain_adaptation}.
