Global Variational Inference Enhanced Robust Domain Adaptation

Abstract

Deep learning-based domain adaptation (DA) methods have shown strong performance by learning transferable representations. However, their reliance on mini-batch training limits global distribution modeling, leading to unstable alignment and suboptimal generalization. We propose Global Variational Inference Enhanced Domain Adaptation (GVI-DA), a framework that learns continuous, class-conditional global priors via variational inference to enable structure-aware cross-domain alignment. GVI-DA minimizes domain gaps through latent feature reconstruction, and mitigates posterior collapse using global codebook learning with randomized sampling. It further improves robustness by discarding low-confidence pseudo-labels and generating reliable target-domain samples. Extensive experiments on four benchmarks and thirty-eight DA tasks demonstrate consistent state-of-the-art performance. We also derive the model's evidence lower bound (ELBO) and analyze the effects of prior continuity, codebook size, and pseudo-label noise tolerance. In addition, we compare GVI-DA with diffusion-based generative frameworks in terms of optimization principles and efficiency, highlighting both its theoretical soundness and practical advantages.

The current version has issues in experimental protocol and presentation. Some evaluation settings (Office-Home/ImageCLEF splits & baselines; Secs.4.1 -- 4.2; Tabs.3 -- 5) are not fully aligned with recent practice, and several figures have labeling/flow issues (e.g., Fig.1(e), Figs.2 -- 4). A revised version will follow