Global Variational Inference Enhanced Robust Domain Adaptation

TL;DR

This paper introduces GVI-DA, a novel domain adaptation framework that leverages global variational inference to improve cross-domain alignment, robustness, and generalization in deep learning models.

Contribution

GVI-DA is the first to incorporate global variational inference with class-conditional priors for structure-aware domain adaptation, enhancing stability and performance.

Findings

Achieves state-of-the-art results on four benchmarks and thirty-eight tasks.

Effectively mitigates pseudo-label noise and improves robustness.

Provides theoretical analysis of ELBO and model components.

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.