A3: Active Adversarial Alignment for Source-Free Domain Adaptation
This work addresses domain adaptation challenges for machine learning applications in scenarios where source data is inaccessible, offering a robust solution with incremental improvements over existing methods.
The paper tackles the problem of source-free unsupervised domain adaptation, where only unlabeled target data is available, by proposing A3, a framework that combines self-supervised learning, adversarial training, and active learning, resulting in improved performance on benchmarks like Office-31 and VisDA-2017 with accuracy gains of up to 5.2%.
Unsupervised domain adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target domain. Recent works have focused on source-free UDA, where only target data is available. This is challenging as models rely on noisy pseudo-labels and struggle with distribution shifts. We propose Active Adversarial Alignment (A3), a novel framework combining self-supervised learning, adversarial training, and active learning for robust source-free UDA. A3 actively samples informative and diverse data using an acquisition function for training. It adapts models via adversarial losses and consistency regularization, aligning distributions without source data access. A3 advances source-free UDA through its synergistic integration of active and adversarial learning for effective domain alignment and noise reduction.