Abstract
Although most transfer learning methods can reduce the difference of the feature distributions between the source and target domains effectively, some classes in the two domains may still be misaligned after domain adaptation, especially for the classes with similar features such as “bicycle” and “motorcycle”. Therefore, a graph regularization based adversarial network model is proposed, whose innovations mainly include the following two aspects: First, a constraint function which is used to measure the difference between the features belonged to different classes is proposed, whose purpose is that not only the training accuracy is taken into account during supervised training, but also the difference between classes should be enlarged as much as possible; Then, a graph regularization constraint function is proposed, which makes all the classes have good local preserving properties after domain adaptation, and further reduces the possibility of all classes being misaligned. Experimental results on several cross-domain benchmark datasets show that our newly proposed approach outperforms state of the art methods.
Highlights
For image classification, most of traditional methods need a large number of labeled samples during training, and obtain the recognition model through supervised learning
We propose a new function to measure the discrimination between classes, all classes are strongly distinguishable from each other through imposing the discrimination constraint on the model as shown in Fig. 1; 2) To align all the classes in the source and target domains accurately, a graph regularization constraint is imposed on the feature layer in the proposed model, all the classes have good locality preserving properties after domain adaptation
We propose a new graph regularization based domain adaptation model, which can improve the discrimination of features between classes, and make the feature distributions of all classes in the source and target domains similar sufficiently
Summary
Most of traditional methods need a large number of labeled samples during training, and obtain the recognition model through supervised learning. As we know, when there is a certain difference in feature distributions between two datasets, the knowledge can be still transferred from one dataset to the other dataset by domain adaptation, i.e., from the source domain to the target domain, so for small sample learning it is very necessary to propose an effective domain adaptation method [1] It can be seen from the existing researches on domain adaptation that the most prominent problem we face is that some classes with strong similarity in the source and target domains are likely to be misaligned after domain adaptation, and the samples which belong to these classes in the target domain may be misclassified [2].
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.