Conformal Risk Minimization for Semi-Supervised Domain Adaptation via Optimal Transport
2026-08-24 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address a challenge in healthcare where machine learning models trained on one patient group don’t work well on another because the data is different. They combine semi-supervised domain adaptation (SSDA) with conformal prediction (CP) to build models that are not only accurate but also give trustworthy uncertainty estimates. By using a technique called Optimal Transport to create labels for unlabeled patient data, their method improves training when only a small amount of labeled data is available. This results in smarter models that make reliable predictions and avoid conflicting diagnoses, especially in tasks like skin lesion classification.
Semi-Supervised Domain AdaptationDistribution ShiftConformal PredictionConformal Risk MinimizationOptimal TransportUncertainty QuantificationNonconformity ScorePrediction SetsDomain InvarianceSkin Lesion Classification
Authors
Manos Giannopoulos, Yi Shen, Michael M. Zavlanos
Abstract
In high-stakes healthcare applications, machine learning models are frequently trained on data from one patient population and deployed on another, creating a distribution shift that degrades both accuracy and reliability. Semi-Supervised Domain Adaptation (SSDA) addresses this by leveraging labeled data from some source domain to improve model performance on a target domain where labels are scarce. However, existing SSDA methods optimize primarily for point-prediction accuracy and offer no principled uncertainty quantification --- a prerequisite for clinical trust. Conformal Prediction (CP) can address this limitation by providing prediction sets with rigorous, distribution-free coverage guarantees. However, applying CP post-hoc to a pre-trained model can yield prohibitively large prediction sets, as SSDA pre-training methods do not account for the nonconformity score geometry that determines conformal set size. Conformal Risk Minimization (CRM) has been used to resolve this issue in the fully supervised setting by integrating the CP objective directly into model training, but it requires a large labeled dataset to compute nonconformity thresholds during training, precisely the data that is scarce in the SSDA regime. We propose an end-to-end framework that integrates CRM into the SSDA training objective, enabling effective CRM in the limited-labeled-target-data regime. The key idea is to utilize Optimal Transport (OT) to generate pseudolabels for unlabeled target instances, providing the additional training signal needed by CRM to operate using only a small labeled target set. This results in a model jointly optimized for domain invariance and conformal efficiency, producing prediction sets that are compact, coverage-valid, and support domain-specific constraints such as excluding mutually contradictory diagnoses in skin lesion classification.