Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence
This is an incremental improvement for domain adaptation methods that use pseudo labeling.
The paper tackles the problem of domain adaptation by proposing to use the discriminator for both learning domain-invariant representations and estimating pseudo labeling confidence, aiming to improve classifier generalization to unlabeled target data.
Often domain adaptation is performed using a discriminator (domain classifier) to learn domain-invariant feature representations so that a classifier trained on labeled source data will generalize well to unlabeled target data. A line of research stemming from semi-supervised learning uses pseudo labeling to directly generate "pseudo labels" for the unlabeled target data and trains a classifier on the now-labeled target data, where the samples are selected or weighted based on some measure of confidence. In this paper, we propose multi-purposing the discriminator to not only aid in producing domain-invariant representations but also to provide pseudo labeling confidence.