Domain Adaptation with Conditional Transferable Components
Domain Adaptation with Conditional Transferable Components
复制标题
DOI：
--
复制
发表时间：
2016-06
期刊：
影响因子：
--
通讯作者：
Mingming Gong;Kun Zhang;Tongliang Liu;D. Tao;C. Glymour;B. Scholkopf
中科院分区：
文献类型：
--
作者：
Mingming Gong;Kun Zhang;Tongliang Liu;D. Tao;C. Glymour;B. Scholkopf
在监督学习中，当训练（源域）和测试（目标域）数据具有不同的分布时，就会出现域自适应。设X和Y分别表示特征和目标，以前关于域自适应的工作主要考虑协变量移位的情况，其中特征P（X）的分布在域之间变化，而条件分布P（Y <$X）保持不变。为了减少域差异，最近的方法试图通过显式地最小化分布差异度量来找到在不同域上具有相似[公式：见文本]的不变分量[公式：见文本]。然而，当P（Y <$X）变化时，不同域中的[公式：见正文]是否也相似尚不清楚。此外，可转移成分不一定是不变的。如果某些成分的变化是可识别的，我们可以利用这些成分在目标域中进行预测。在本文中，我们关注的情况下，P（X <$Y）和P（Y）都改变的因果系统，其中Y是X的原因。在适当的假设下，我们的目标是提取条件可转移成分，其条件分布[公式：见正文]在适当的位置-尺度（LS）变换后不变，并同时识别P（Y）如何在域之间变化。我们提供了理论分析和实证评估的合成和现实世界的数据，以显示我们的方法的有效性。
Domain adaptation arises in supervised learning when the training (source domain) and test (target domain) data have different distributions. Let X and Y denote the features and target, respectively, previous work on domain adaptation mainly considers the covariate shift situation where the distribution of the features P(X) changes across domains while the conditional distribution P(Y∣X) stays the same. To reduce domain discrepancy, recent methods try to find invariant components [Formula: see text] that have similar [Formula: see text] on different domains by explicitly minimizing a distribution discrepancy measure. However, it is not clear if [Formula: see text] in different domains is also similar when P(Y∣X) changes. Furthermore, transferable components do not necessarily have to be invariant. If the change in some components is identifiable, we can make use of such components for prediction in the target domain. In this paper, we focus on the case where P(X∣Y) and P(Y) both change in a causal system in which Y is the cause for X. Under appropriate assumptions, we aim to extract conditional transferable components whose conditional distribution [Formula: see text] is invariant after proper location-scale (LS) transformations, and identify how P(Y) changes between domains simultaneously. We provide theoretical analysis and empirical evaluation on both synthetic and real-world data to show the effectiveness of our method.