Target contrastive pessimistic risk for robust domain adaptation
This addresses the robustness issue in domain adaptation for machine learning applications, though it appears incremental as it builds on existing methods with a conservative approach.
The paper tackles the problem of domain adaptation where adaptive classifiers can underperform due to invalid assumptions, by developing a robust classifier that avoids restrictive assumptions and ensures it does not perform worse than a non-adaptive one. The result shows that their method performs on par with state-of-the-art classifiers in sample selection bias settings and outperforms them in more general domain adaptation settings.
In domain adaptation, classifiers with information from a source domain adapt to generalize to a target domain. However, an adaptive classifier can perform worse than a non-adaptive classifier due to invalid assumptions, increased sensitivity to estimation errors or model misspecification. Our goal is to develop a domain-adaptive classifier that is robust in the sense that it does not rely on restrictive assumptions on how the source and target domains relate to each other and that it does not perform worse than the non-adaptive classifier. We formulate a conservative parameter estimator that only deviates from the source classifier when a lower risk is guaranteed for all possible labellings of the given target samples. We derive the classical least-squares and discriminant analysis cases and show that these perform on par with state-of-the-art domain adaptive classifiers in sample selection bias settings, while outperforming them in more general domain adaptation settings.