Abstract:Unsupervised multi-source domain adaptation (UMDA) leverages labeled data from multiple source domains to generalize to an unlabeled target. While federated UMDA addresses privacy by avoiding raw data sharing, existing methods scale poorly as the number of sources increases, often suffering from high computational overhead or training instability. We propose GALA, a scalable and robust federated UMDA framework designed for high-diversity settings. GALA achieves scalability by coupling a novel inter-group discrepancy minimization objective that approximates pairwise alignment with linear complexity alongside a temperature-controlled, centroid-based weighting strategy for dynamic source prioritization. These components enable stable, parallelizable training across many heterogeneous sources, addressing a critical scalability bottleneck that remains largely unaddressed in current literature. To evaluate performance in high-diversity scenarios, we introduce Digit-18, a new benchmark comprising 18 datasets with varied synthetic and real-world domain shifts. Extensive experiments demonstrate that GALA achieves state-of-the-art results on standard benchmarks and significantly outperforms prior methods in large-scale settings where others either fail to converge or become computationally infeasible.
| Comments: | Accepted at ICML 2026. Pre-camera-ready version |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2510.08150 [cs.LG] |
| (or arXiv:2510.08150v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2510.08150 arXiv-issued DOI via DataCite |
From: Cem Ata Baykara [view email]
[v1]
Thu, 9 Oct 2025 12:34:37 UTC (1,087 KB)
[v2]
Fri, 10 Oct 2025 08:35:27 UTC (1,087 KB)
[v3]
Tue, 5 May 2026 14:15:16 UTC (1,032 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。