Conformal Risk-Averse Decision Making with Action Conditional Guarantees

ICLR 2026 Conference Submission13415 Authors

Keywords: Conformal prediction, Uncertainty quantification, Risk averse, Risk sensitive, Decision making, Safety guarantees

Abstract: Reliable decision-making pipelines powered by machine learning models require uncertainty quantification (UQ) methods that come with explicit safety guarantees. Conformal prediction provides such UQ by wrapping ML predictions into prediction sets, and recent work has shown that these sets can be translated into optimal risk-averse decision policies—though only under marginal safety guarantees [Kiyani et al., 2025]. We generalize and strengthen this connection by (i) introducing action-conditional conformal prediction, which yields safety guarantees conditioned explicitly on each action taken by the decision maker, (ii) proving that action-conditional prediction sets characterize the feasible decision space for risk-averse decision makers aiming to optimize action-conditional value-at-risk, and (iii) proposing a principled finite-sample algorithm based on pinball-loss minimization that extends the conditional validity framework of [Gibbs et al., 2025] to the action-conditional setting. Experiments on two real-world datasets confirm that our approach significantly improves action-conditional performance over prior work.

Primary Area: probabilistic methods (Bayesian methods, variational inference, sampling, UQ, etc.)

Submission Number: 13415