Machine learning models are often seen as black-box systems that output point predictions without indicating how confident they are in those predictions. Recently, Conformal Prediction (CP) has emerged as a powerful framework to address this issue by transforming point predictions into set-valued predictions with probabilistic guarantees. In this talk, I will introduce CP and briefly present some key challenges and recent advances in the area. I will first discuss how to perform CP in a Federated Learning setting, showing that a single round of communication is sufficient to match the performance of centralized approaches. I will then turn to the question of efficiency control in CP, where finite-sample guarantees can be obtained by viewing CP as a minimum-volume set estimation problem.

This talk is based on the following papers:

P. Humbert, B. Le Bars, A. Bellet, S. Arlot. One-Shot Federated Conformal Prediction, ICML 2023

P. Humbert, B. Le Bars, A. Bellet, S. Arlot. Marginal and Training-Conditional Guarantees in One-Shot Federated Conformal Prediction, Arxiv preprint 2024

B. Le Bars, P. Humbert. On Volume Minimization in Conformal Regression, ICML 2025