![[2602.23296] Conformalized Neural Networks for Federated Uncertainty Quantification under Dual Heterogeneity](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.23296] Conformalized Neural Networks for Federated Uncertainty Quantification under Dual Heterogeneity
Summary
This article presents FedWQ-CP, a novel approach to federated uncertainty quantification that addresses dual heterogeneity in data and model performance, enhancing reliability in federated learning systems.
Why It Matters
As federated learning becomes increasingly prevalent, ensuring the reliability of uncertainty quantification is crucial for deploying effective AI models. This research addresses a significant gap in existing methods by integrating data and model heterogeneity, which can lead to better performance and reduced risks of local failures in AI applications.
Key Takeaways
- FedWQ-CP balances empirical coverage performance with efficiency in federated learning.
- The approach allows for agent-server calibration in a single communication round.
- It maintains both agent-wise and global coverage while minimizing prediction intervals.
- Experimental results demonstrate effectiveness across various datasets.
- Addresses the joint effects of data and model heterogeneity in uncertainty quantification.
Computer Science > Machine Learning arXiv:2602.23296 (cs) [Submitted on 26 Feb 2026] Title:Conformalized Neural Networks for Federated Uncertainty Quantification under Dual Heterogeneity Authors:Quang-Huy Nguyen, Jiaqi Wang, Wei-Shinn Ku View a PDF of the paper titled Conformalized Neural Networks for Federated Uncertainty Quantification under Dual Heterogeneity, by Quang-Huy Nguyen and Jiaqi Wang and Wei-Shinn Ku View PDF HTML (experimental) Abstract:Federated learning (FL) faces challenges in uncertainty quantification (UQ). Without reliable UQ, FL systems risk deploying overconfident models at under-resourced agents, leading to silent local failures despite seemingly satisfactory global performance. Existing federated UQ approaches often address data heterogeneity or model heterogeneity in isolation, overlooking their joint effect on coverage reliability across agents. Conformal prediction is a widely used distribution-free UQ framework, yet its applications in heterogeneous FL settings remains underexplored. We provide FedWQ-CP, a simple yet effective approach that balances empirical coverage performance with efficiency at both global and agent levels under the dual heterogeneity. FedWQ-CP performs agent-server calibration in a single communication round. On each agent, conformity scores are computed on calibration data and a local quantile threshold is derived. Each agent then transmits only its quantile threshold and calibration sample size to the server. The server ...
