Divide and Adapt: Active Domain Adaptation via Customized Learning

TL;DR

This paper introduces DiaNA, a novel active domain adaptation framework that partitions target data into categories based on uncertainty and domainness, enabling tailored learning strategies to improve adaptation across various settings.

Contribution

The paper proposes a new data subdivision protocol and informativeness score for active domain adaptation, effectively recognizing valuable samples and handling large domain gaps.

Findings

DiaNA accurately identifies informative samples for annotation.

It generalizes well across different domain adaptation settings.

The framework improves adaptation performance with stratified data handling.

Abstract

Active domain adaptation (ADA) aims to improve the model adaptation performance by incorporating active learning (AL) techniques to label a maximally-informative subset of target samples. Conventional AL methods do not consider the existence of domain shift, and hence, fail to identify the truly valuable samples in the context of domain adaptation. To accommodate active learning and domain adaption, the two naturally different tasks, in a collaborative framework, we advocate that a customized learning strategy for the target data is the key to the success of ADA solutions. We present Divide-and-Adapt (DiaNA), a new ADA framework that partitions the target instances into four categories with stratified transferable properties. With a novel data subdivision protocol based on uncertainty and domainness, DiaNA can accurately recognize the most gainful samples. While sending the informative instances for annotation, DiaNA employs tailored learning strategies for the remaining categories. Furthermore, we propose an informativeness score that unifies the data partitioning criteria. This enables the use of a Gaussian mixture model (GMM) to automatically sample unlabeled data into the proposed four categories. Thanks to the "divideand-adapt" spirit, DiaNA can handle data with large variations of domain gap. In addition, we show that DiaNA can generalize to different domain adaptation settings, such as unsupervised domain adaptation (UDA), semi-supervised domain adaptation (SSDA), source-free domain adaptation (SFDA), etc.