S$^3$VAADA: Submodular Subset Selection for Virtual Adversarial Active Domain Adaptation
This addresses the challenge of efficiently combining limited labeled target data with source knowledge for improved generalization in domain adaptation, though it appears incremental as it builds on existing cluster-based methods.
The paper tackles the problem of selecting maximally-informative samples to label in active domain adaptation, where only a small proportion of target data can be labeled, and proposes S^3VAADA, which consistently outperforms state-of-the-art approaches on datasets with varying domain shifts.
Unsupervised domain adaptation (DA) methods have focused on achieving maximal performance through aligning features from source and target domains without using labeled data in the target domain. Whereas, in the real-world scenario's it might be feasible to get labels for a small proportion of target data. In these scenarios, it is important to select maximally-informative samples to label and find an effective way to combine them with the existing knowledge from source data. Towards achieving this, we propose S$^3$VAADA which i) introduces a novel submodular criterion to select a maximally informative subset to label and ii) enhances a cluster-based DA procedure through novel improvements to effectively utilize all the available data for improving generalization on target. Our approach consistently outperforms the competing state-of-the-art approaches on datasets with varying degrees of domain shifts.