Domain Generalization via Selective Consistency Regularization for Time Series Classification

TL;DR

This paper introduces a novel domain generalization approach for time-series classification that selectively enforces consistency between closely-related source domains, improving robustness and performance without target domain data.

Contribution

It proposes a new representation learning method that avoids negative transfer by focusing on related domains, enhancing domain generalization in time-series tasks.

Findings

Significantly outperforms baseline methods.

Achieves better or competitive accuracy compared to state-of-the-art.

Improves model calibration in domain generalization.

Abstract

Domain generalization methods aim to learn models robust to domain shift with data from a limited number of source domains and without access to target domain samples during training. Popular domain alignment methods for domain generalization seek to extract domain-invariant features by minimizing the discrepancy between feature distributions across all domains, disregarding inter-domain relationships. In this paper, we instead propose a novel representation learning methodology that selectively enforces prediction consistency between source domains estimated to be closely-related. Specifically, we hypothesize that domains share different class-informative representations, so instead of aligning all domains which can cause negative transfer, we only regularize the discrepancy between closely-related domains. We apply our method to time-series classification tasks and conduct comprehensive experiments on three public real-world datasets. Our method significantly improves over the baseline and achieves better or competitive performance in comparison with state-of-the-art methods in terms of both accuracy and model calibration.