TWINs: Two Weighted Inconsistency-reduced Networks for Partial Domain Adaptation
This addresses the problem of domain adaptation in scenarios with mismatched class sets, which is incremental as it builds on prior PDA methods.
The paper tackles partial domain adaptation (PDA), where target classes are a subset of source classes, by proposing TWINs, a method that uses two weighted classification networks to estimate target class ratios and minimize classifier inconsistency, achieving large-margin performance gains over existing methods on several datasets.
The task of unsupervised domain adaptation is proposed to transfer the knowledge of a label-rich domain (source domain) to a label-scarce domain (target domain). Matching feature distributions between different domains is a widely applied method for the aforementioned task. However, the method does not perform well when classes in the two domains are not identical. Specifically, when the classes of the target correspond to a subset of those of the source, target samples can be incorrectly aligned with the classes that exist only in the source. This problem setting is termed as partial domain adaptation (PDA). In this study, we propose a novel method called Two Weighted Inconsistency-reduced Networks (TWINs) for PDA. We utilize two classification networks to estimate the ratio of the target samples in each class with which a classification loss is weighted to adapt the classes present in the target domain. Furthermore, to extract discriminative features for the target, we propose to minimize the divergence between domains measured by the classifiers' inconsistency on target samples. We empirically demonstrate that reducing the inconsistency between two networks is effective for PDA and that our method outperforms other existing methods with a large margin in several datasets.