Implicit Class-Conditioned Domain Alignment for Unsupervised Domain Adaptation
This work addresses domain adaptation challenges for machine learning applications where labeled data is scarce, but it appears incremental as it builds on existing alignment methods.
The paper tackles unsupervised domain adaptation by addressing pseudo-label bias in class-conditioned alignment, proposing an implicit sampling-based method that improves performance under class imbalance and distribution shift, with empirical results confirming its effectiveness.
We present an approach for unsupervised domain adaptation---with a strong focus on practical considerations of within-domain class imbalance and between-domain class distribution shift---from a class-conditioned domain alignment perspective. Current methods for class-conditioned domain alignment aim to explicitly minimize a loss function based on pseudo-label estimations of the target domain. However, these methods suffer from pseudo-label bias in the form of error accumulation. We propose a method that removes the need for explicit optimization of model parameters from pseudo-labels directly. Instead, we present a sampling-based implicit alignment approach, where the sample selection procedure is implicitly guided by the pseudo-labels. Theoretical analysis reveals the existence of a domain-discriminator shortcut in misaligned classes, which is addressed by the proposed implicit alignment approach to facilitate domain-adversarial learning. Empirical results and ablation studies confirm the effectiveness of the proposed approach, especially in the presence of within-domain class imbalance and between-domain class distribution shift.