Mitigating Uncertainty of Classifier for Unsupervised Domain Adaptation
This work addresses domain adaptation for machine learning applications where labeled data is scarce, but it appears incremental as it builds on existing knowledge by analyzing classifier distributions.
The paper tackles the problem of unsupervised domain adaptation by thoroughly examining the role of classifier performance in matching source and target distributions, specifically investigating feature distributions, probabilistic uncertainty, and certainty activation mappings, resulting in consistently improved performance across all datasets.
Understanding unsupervised domain adaptation has been an important task that has been well explored. However, the wide variety of methods have not analyzed the role of a classifier's performance in detail. In this paper, we thoroughly examine the role of a classifier in terms of matching source and target distributions. We specifically investigate the classifier ability by matching a) the distribution of features, b) probabilistic uncertainty for samples and c) certainty activation mappings. Our analysis suggests that using these three distributions does result in a consistently improved performance on all the datasets. Our work thus extends present knowledge on the role of the various distributions obtained from the classifier towards solving unsupervised domain adaptation.