GALA: A GlobAl-LocAl Approach for Multi-Source Active Domain Adaptation

2025-10-25

AI Summary

This paper addresses the challenge of sample selection in multi-source active domain adaptation (MS-ADA), where concurrent inter-class diversity and multi-source domain shifts impede reliable query selection. To tackle this, we propose GALA—a global-local joint selection strategy that synergistically integrates k-means-based global clustering with intra-cluster uncertainty-driven local selection. GALA introduces a parameter-free, plug-and-play sample selection framework requiring no additional trainable parameters. While maintaining computational efficiency, GALA significantly improves target-labeling efficiency: on three standard benchmarks, it achieves near fully supervised performance using only 1% labeled target data—substantially outperforming existing MS-ADA methods. Its core contribution lies in decoupling the modeling of cross-domain discrepancies from intra-class distribution heterogeneity, establishing a novel paradigm for multi-source active learning.

Technology Category

• Machine Learning: Active Learning
• Multiagent Systems: Adversarial Agents
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Domain Adaptation (DA) provides an effective way to tackle target-domain tasks by leveraging knowledge learned from source domains. Recent studies have extended this paradigm to Multi-Source Domain Adaptation (MSDA), which exploits multiple source domains carrying richer and more diverse transferable information. However, a substantial performance gap still remains between adaptation-based methods and fully supervised learning. In this paper, we explore a more practical and challenging setting, named Multi-Source Active Domain Adaptation (MS-ADA), to further enhance target-domain performance by selectively acquiring annotations from the target domain. The key difficulty of MS-ADA lies in designing selection criteria that can jointly handle inter-class diversity and multi-source domain variation. To address these challenges, we propose a simple yet effective GALA strategy (GALA), which combines a global k-means clustering step for target-domain samples with a cluster-wise local selection criterion, effectively tackling the above two issues in a complementary manner. Our proposed GALA is plug-and-play and can be seamlessly integrated into existing DA frameworks without introducing any additional trainable parameters. Extensive experiments on three standard DA benchmarks demonstrate that GALA consistently outperforms prior active learning and active DA methods, achieving performance comparable to the fully-supervised upperbound while using only 1% of the target annotations.

Problem

• Enhancing target-domain performance with selective annotations
• Addressing inter-class diversity and multi-source domain variation
• Achieving near-supervised performance with minimal target labels

Innovation

• Combines global clustering with local selection
• Plug-and-play integration into existing frameworks
• Uses only 1% target annotations for near-supervised performance