Personalized Federated Conformal Prediction with Localization

Liuhua Peng (University of Melbourne) · Yinjie Min (Nankai University) · Chuchen Zhang (Nankai University) · Changliang Zou

agent-personalized prediction setsconditional coverage performanceconformal prediction data heterogeneity extensive experimentsinstance-localizationmarginal coverage guaranteespersonalized federated conformal predictionpersonalized federated learningprivacy-preserving knowledge transferrisk-sensitive applicationssource agentsstatistical validitytarget agentsuncertainty quantification

Personalized federated learning addresses data heterogeneity across distributed agents but lacks uncertainty quantification that is both agent-specific and instance-specific, which is a critical requirement for risk-sensitive applications. We propose personalized federated conformal prediction (PFCP), a novel framework that combines personalized federated learning with conformal prediction to provide statistically valid agent-personalized prediction sets with instance-localization. By leveraging privacy-preserving knowledge transfer from other source agents, PFCP ensures marginal coverage guarantees for target agents while significantly improving conditional coverage performance on individual test instances, which has been validated by extensive experiments.