Self-corrected unsupervised domain adaptation |
| |
Authors: | Yunyun WANG Chao WANG Hui XUE Songcan CHEN |
| |
Affiliation: | 1. School of Computer Science and Engineering, Nanjing University of Posts & Telecommunications, Nanjing 210046, China2. School of Computer Science and Engineering, Southeast University, Nanjing 210096, China3. School of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210023, China |
| |
Abstract: | Unsupervised domain adaptation (UDA), which aims to use knowledge from a label-rich source domain to help learn unlabeled target domain, has recently attracted much attention. UDA methods mainly concentrate on source classification and distribution alignment between domains to expect the correct target prediction. While in this paper, we attempt to learn the target prediction end to end directly, and develop a Self-corrected unsupervised domain adaptation (SCUDA) method with probabilistic label correction. SCUDA adopts a probabilistic label corrector to learn and correct the target labels directly. Specifically, besides model parameters, those target pseudo-labels are also updated in learning and corrected by the anchor-variable, which preserves the class candidates for samples. Experiments on real datasets show the competitiveness of SCUDA. |
| |
Keywords: | unsupervised domain adaptation adversarial Learning deep neural network pseudo-labels label corrector |
|
| 点击此处可从《Frontiers of Computer Science》浏览原始摘要信息 |
|
点击此处可从《Frontiers of Computer Science》下载全文 |
|