Year: 2024
Type: conference-paper
Abstract: 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. Our approach significantly advances state-of-the-art methods, achieving 4.1% on Office-31, 11.7% on Office-Home, and 10.6% on DomainNet accuracy improvements. Source code: https://github.com/chrisantuseze/active-self-pretraining
Authors: Chrisantus Eze, Christopher Crick
Institution: Oklahoma State University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: National Science Foundation
OpenAlex ID: W4408145758