Partial Model Sharing Improves Byzantine Resilience in Federated Conformal Prediction

TL;DR

This paper introduces a Byzantine-resilient federated conformal prediction method using partial model sharing and histogram-based client evaluation, enhancing robustness and efficiency in federated uncertainty quantification.

Contribution

It presents a novel approach combining partial model sharing with histogram-based client evaluation to improve Byzantine resilience in federated conformal prediction.

Findings

Achieves closer-to-nominal coverage under Byzantine attacks.

Produces substantially tighter prediction intervals than standard FCP.

Reduces communication overhead during federated training.

Abstract

We propose a Byzantine-resilient federated conformal prediction (FCP) method that leverages partial model sharing, where only a subset of model parameters is exchanged each round. Unlike existing robust FCP approaches that primarily harden the calibration stage, our method protects both the federated training and conformal calibration phases. During training, partial sharing inherently restricts the attack surface and attenuates poisoned updates while reducing communication. During calibration, clients compress their non-conformity scores into histogram-based characterization vectors, enabling the server to detect Byzantine clients via distance-based maliciousness scores and to estimate the conformal quantile using only benign contributors. Experiments across diverse Byzantine attack scenarios show that the proposed method achieves closer-to-nominal coverage with substantially tighter prediction intervals than standard FCP, establishing a robust and communication-efficient approach to federated uncertainty quantification.