Blended-target Domain Adaptation
2 papers with code • 3 benchmarks • 3 datasets
Blended-target domain adaptation is to adapt a single source model to multiple different target domains. The task is similar to the multi-target domain adaptation. However, the domain labels are not available.
Most implemented papers
Curriculum Graph Co-Teaching for Multi-Target Domain Adaptation
In this paper we address multi-target domain adaptation (MTDA), where given one labeled source dataset and multiple unlabeled target datasets that differ in data distributions, the task is to learn a robust predictor for all the target domains.
Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation
We demonstrate that domain labels are not directly necessary for BTDA if categorical distributions of various domains are sufficiently aligned even facing the imbalance of domains and the label distribution shift of classes.