Communication-Efficient Byzantine-Robust Federated Conformal Prediction via Partial Model Sharing
arXiv:2602.18396v2 Announce Type: replace Abstract: We propose PRISM-FCP (Partial shaRing and robust calIbration with Statistical Margins for Federated Conformal Prediction), a communication-efficient Byzantine-robust federated conformal prediction framework that uses partial model sharing to mitigate stochastic model-poisoning attacks during training and histogram-based filtering to mitigate adversarial calibration submissions. Existing approaches address adversarial behavior only in the calibration stage, leaving the learned model susceptible to poisoned updates. In contrast, PRISM-FCP mitig
The increasing adoption of federated learning in sensitive applications necessitates robust defenses against Byzantine attacks, which are becoming more sophisticated.
Ensuring the security and reliability of federated AI models is crucial for their deployment in critical infrastructure and privacy-sensitive domains.
This research introduces a framework that significantly enhances communication efficiency and Byzantine robustness in federated learning, moving beyond existing calibration-stage-only defenses.
- · Organizations deploying federated learning models
- · Cybersecurity sector
- · AI fairness and safety researchers
- · Actors involved in AI model poisoning attacks
- · Less secure federated learning frameworks
More secure and trustworthy federated AI models will be deployed across industries.
Increased adoption of federated learning could accelerate innovation in privacy-preserving AI applications.
The development of more resilient AI systems may lead to new regulatory standards for AI model security and robustness.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG