Two-Stage Alignments Framework for Unsupervised Domain Adaptation on Time Series Data

Unsupervised Domain Adaptation (UDA) aims to free models from labeled information of target domain by minimizing the discrepancy of distributions between different domains. Most existing methods are designed to learn domain-invariant features either by domain discrimination or by matching lower-order moments. However, these methods are not robust due to the limited representation of statistical characteristics for non-Gaussian distributions and thus fail in domain matching. In addition, they often focus on matching distributions while not considering class decision boundaries between domains. To address these issues, we propose a novel Two-Stage Alignments Framework (TSAF) for UAD, which not only performs arbitrary-order moment matching to approximately characterize complex non-Gaussian distributions, but also utilizes domain-specific decision boundaries to align the probabilistic outputs of classifiers. Moreover, the reconstruction-based task is introduced to enhance the representation of the inherent characteristics for specific distribution. Extensive experiments on three real-world time series datasets demonstrate that: 1) our model evidently outperforms many state-of-the-art domain adaptation methods in cross-domain classification tasks; 2) TSAF can learn domain-invariant features efficiently.

Automated summary

Unsupervised Domain Adaptation (UDA) aims to free models from labeled information in the target domain and minimize the distribution discrepancies between different domains. Most existing methods focus only on domain-invariant feature learning through either domain discrimination or matching lower-order moments, resulting in limited robustness for non-Gaussian distributions and inadequate domain matching. To address these issues, we propose a novel Two-Stage Alignments Framework (TSAF) for UDA, which characterizes non-Gaussian distributions through arbitrary-order moment matching and aligns probabilistic outputs of classifiers using domain-specific decision boundaries. Additionally, a reconstruction-based task is introduced to enhance the representation of specific distribution characteristics. Experimental results on three real-world time series datasets demonstrate the superiority of our model in cross-domain classification tasks and the efficient learning of domain-invariant features by TSAF.