Learning Semantic Representations for Unsupervised Domain Adaptation

ICML 2018  ·  Shaoan Xie, Zibin Zheng, Liang Chen, Chuan Chen ·

It is important to transfer the knowledge from label-rich source domain to unlabeled target domain due to the expensive cost of manual labeling efforts. Prior domain adaptation methods address this problem through aligning the global distribution statistics between source domain and target domain, but a drawback of prior methods is that they ignore the semantic information contained in samples, e.g., features of backpacks in target domain might be mapped near features of cars in source domain. In this paper, we present moving semantic transfer network, which learn semantic representations for unlabeled target samples by aligning labeled source centroid and pseudo-labeled target centroid. Features in same class but different domains are expected to be mapped nearby, resulting in an improved target classification accuracy. Moving average centroid alignment is cautiously designed to compensate the insufficient categorical information within each mini batch. Experiments testify that our model yields state of the art results on standard datasets.

PDF Abstract

Results from the Paper

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Source Paper Compare
Domain Adaptation SVHN-to-MNIST MSTN Accuracy 93.3 # 8


No methods listed for this paper. Add relevant methods here