Efficient Federated Conformal Prediction with Group-Conditional Guarantee

Haifeng Wen, Osvaldo Simeone, Hong Xing
Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, PMLR 337:7290-7313, 2026.

Abstract

Deploying trustworthy {AI} systems requires principled uncertainty quantification. Conformal prediction ({CP}) is a widely used framework for constructing prediction sets with distribution-free coverage guarantees. In many practical settings, including healthcare, finance, and mobile sensing, the calibration data required for {CP} are distributed across multiple clients, each with its own local data distribution. In this federated setting, data can often be partitioned into, potentially overlapping, groups, which may reflect client-specific strata or cross-cutting attributes such as demographic or semantic categories. We propose \emph{group-conditional} federated conformal prediction (GC-FCP), a federated extension of conditional conformal calibration for a target mixture over prespecified groups. GC-FCP constructs mergeable, atom-stratified coresets from local calibration scores, enabling compact aggregation at the server when the number of active atoms is moderate. Experiments on synthetic and real-world datasets validate the performance of GC-FCP compared to centralized calibration baselines. The code of our work can be found at https://github.com/HaifengWen/GC-FCP.

Cite this Paper

BibTeX
@InProceedings{pmlr-v337-wen26a, title = {Efficient Federated Conformal Prediction with Group-Conditional Guarantee}, author = {Wen, Haifeng and Simeone, Osvaldo and Xing, Hong}, booktitle = {Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence}, pages = {7290--7313}, year = {2026}, editor = {Perković, Emilija and Malinsky, Daniel}, volume = {337}, series = {Proceedings of Machine Learning Research}, month = {17--21 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v337/main/assets/wen26a/wen26a.pdf}, url = {https://proceedings.mlr.press/v337/wen26a.html}, abstract = {Deploying trustworthy {AI} systems requires principled uncertainty quantification. Conformal prediction ({CP}) is a widely used framework for constructing prediction sets with distribution-free coverage guarantees. In many practical settings, including healthcare, finance, and mobile sensing, the calibration data required for {CP} are distributed across multiple clients, each with its own local data distribution. In this federated setting, data can often be partitioned into, potentially overlapping, groups, which may reflect client-specific strata or cross-cutting attributes such as demographic or semantic categories. We propose \emph{group-conditional} federated conformal prediction (GC-FCP), a federated extension of conditional conformal calibration for a target mixture over prespecified groups. GC-FCP constructs mergeable, atom-stratified coresets from local calibration scores, enabling compact aggregation at the server when the number of active atoms is moderate. Experiments on synthetic and real-world datasets validate the performance of GC-FCP compared to centralized calibration baselines. The code of our work can be found at https://github.com/HaifengWen/GC-FCP.} }
Endnote
%0 Conference Paper %T Efficient Federated Conformal Prediction with Group-Conditional Guarantee %A Haifeng Wen %A Osvaldo Simeone %A Hong Xing %B Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2026 %E Emilija Perković %E Daniel Malinsky %F pmlr-v337-wen26a %I PMLR %P 7290--7313 %U https://proceedings.mlr.press/v337/wen26a.html %V 337 %X Deploying trustworthy {AI} systems requires principled uncertainty quantification. Conformal prediction ({CP}) is a widely used framework for constructing prediction sets with distribution-free coverage guarantees. In many practical settings, including healthcare, finance, and mobile sensing, the calibration data required for {CP} are distributed across multiple clients, each with its own local data distribution. In this federated setting, data can often be partitioned into, potentially overlapping, groups, which may reflect client-specific strata or cross-cutting attributes such as demographic or semantic categories. We propose \emph{group-conditional} federated conformal prediction (GC-FCP), a federated extension of conditional conformal calibration for a target mixture over prespecified groups. GC-FCP constructs mergeable, atom-stratified coresets from local calibration scores, enabling compact aggregation at the server when the number of active atoms is moderate. Experiments on synthetic and real-world datasets validate the performance of GC-FCP compared to centralized calibration baselines. The code of our work can be found at https://github.com/HaifengWen/GC-FCP.
APA
Wen, H., Simeone, O. & Xing, H.. (2026). Efficient Federated Conformal Prediction with Group-Conditional Guarantee. Proceedings of the 42nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 337:7290-7313 Available from https://proceedings.mlr.press/v337/wen26a.html.

Related Material