Quantifying uncertainty in clinical predictions is critical for high-stakes diagnosis tasks. Conformal prediction offers a principled approach by providing prediction sets with theoretical coverage guarantees. However, in practice, patient distribution shifts violate the i.i.d. assumptions underlying standard conformal methods, leading to poor coverage in healthcare settings. In this work, we evaluate several conformal prediction approaches on EEG seizure classification, a task with known distribution shift challenges and label uncertainty. We demonstrate that personalized calibration strategies can improve coverage by over 20 percentage points while maintaining comparable prediction set sizes. Our implementation is available via PyHealth, an open-source healthcare AI framework: https://github.com/sunlabuiuc/PyHealth.
Figures & tables
Figure 1: EEG Classification Challenges. EEG tasks often violate key aspects of a typical machine learning pipeline. (a) Their annotation process consistently leaves room for uncertainty, which is then passed onto the models. (b) Training, validation, and test distributions are not i.i.d. due to patient distribution shift [ 13 ] . Both issues make EEG classification a very challenging machine learning problem.
Figure 2: Empirical coverage under random (top) and patient (bottom) splits. The dotted line is target coverage 1−α . Under the random split, NCP substantially outperforms non-personalized baselines on TUEV. Under the patient split, all methods fall short of target coverage, reflecting the difficulty of cross-patient distribution shift.
Figure 3: Average prediction set sizes under random (top) and patient (bottom) splits. NCP consistently maintains smaller prediction sets than non-personalized methods despite achieving competitive or higher coverage.
Figure 4: Empirical coverage vs. α for NCP using varying calibration set sizes k . Shaded regions denote ±1 std. The dashed line is target coverage 1−α .
We consider the problem of uncertainty quantification for a pretrained classification model deployed under unknown distribution shift. We propose Audited Conformal Prediction (ACP), a method that leverages a small labeled dataset from the target population to train an auxiliary audit model identifying inputs where the legacy model is likely to fail. By integrating the audit model's outputs into the conformal prediction framework, ACP produces prediction sets that guarantee marginal coverage while achieving substantially higher conditional coverage in practice than existing approaches. We develop and analyze two complementary integration strategies -- one targeting marginal coverage with improved conditional performance, the other providing explicit group-conditional coverage guarantees -- and establish theoretical guarantees for both. Experiments on synthetic and real-world datasets validate the method and illustrate trade-offs between prediction set size and conditional coverage.
Yanfei Zhou, Rizal Fathony, Nam H. Nguyen +1
Department of Data Sciences and Operations University of Southern California Los Angeles, CA 90089 · AI Foundations, Capital One New York, NY 10017 · Department of Data Sciences and Operations Thomas Lord Department of Computer Science University of Southern California Los Angeles, CA 90089
Conformal prediction (CP) is a widely used frequentist framework to quantify uncertainty by constructing prediction sets with user-specified marginal coverage guarantees. In practice, CP is typically applied on top of probabilistic classifiers, which are able to express aleatoric but not epistemic uncertainty. In this paper, we consider the question of how to optimally employ CP on top of a more expressive formalism, namely credal sets, which can express both aleatoric and epistemic uncertainty. More specifically, we propose probabilistic Bernoulli prediction sets (BPS) and derive a variant that achieves conditional coverage for valid credal sets while remaining minimal in expected size. We then address the more realistic scenario in which the validity of the credal sets is not guaranteed. Assuming access to calibration data with ground-truth distributions over labels, we apply conformal risk control to BPS and derive a PAC-style guarantee: with high probability over the data, the achieved conditional coverage is at least the desired level. We validate our theoretical findings empirically over various datasets.
Alireza Javanmardi, Soroush H. Zargarbashi, Santo M. A. R. Thies +3
Conformal prediction (CP) offers distribution-free marginal coverage guarantees under an exchangeability assumption, but these guarantees can fail if the data distribution shifts. We analyze the use of pseudo-calibration as a tool to counter this performance loss under a bounded label-conditional covariate shift model. Using tools from domain adaptation, we derive a lower bound on target coverage in terms of the source-domain loss of the classifier and a Wasserstein measure of the shift. Using this result, we provide a method to design pseudo-calibrated sets that inflate the conformal threshold by a slack parameter to keep target coverage above a prescribed level. Finally, we propose a source-tuned pseudo-calibration algorithm that interpolates between hard pseudo-labels and randomized labels as a function of classifier uncertainty. Numerical experiments show that our bounds qualitatively track pseudo-calibration behavior and that the source-tuned scheme mitigates coverage degradation under distribution shift while maintaining nontrivial prediction set sizes.
Farbod Siahkali, Ashwin Verma, Vijay Gupta
Elmore Family School of Electrical and Computer Engineering, at Purdue University, West Lafayette, IN 47907 USA