Conformal Risk-Averse Decision Making with Optimized Certainty Equivalent Risk Control
Abstract
We study risk-averse decision making, in which an agent selects actions while being uncertain about the true system state. The risk is measured via optimized certainty equivalent (OCE) metrics, which generalize popular criteria such as mean-variance risk and conditional value-at-risk (CVaR). We characterize the optimal policy under known distributions, and show that it reduces to a prediction set-based solution for the CVaR. This provides an operational interpretation of conformal prediction-type prediction sets. For unknown distributions, we develop a data-driven calibration strategy, based on a synthetic model for the likelihood and held-out calibration data, yielding high-probability control of the OCE risk. The approach is evaluated on two wireless beamforming settings.
Related
- Provably Efficient Risk-Sensitive Offline Reinforcement Learning via Optimized Certainty Equivalents
- Dual Effect and Certainty Equivalence Revisited for a Linear-Gaussian Dual Control Problem
- Selecting Informative Conformal Prediction Sets with an Optimized FCR-Controlled Approach
- CASCADE Conformal Prediction: Uncertainty-Adaptive Prediction Intervals for Two-Stage Clinical Decision Support
- Conformal Robustness in Prediction-Driven Decision-Making
BibTeX
@misc{farzaneh2026conformal,
title = {{Conformal Risk-Averse Decision Making with Optimized Certainty Equivalent Risk Control}},
author = {Amirmohammad Farzaneh and Osvaldo Simeone},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.28179},
eprint = {2608.28179},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.28179},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Conformal Risk-Averse Decision Making with Optimized Certainty Equivalent Risk Control
AU - Amirmohammad Farzaneh
AU - Osvaldo Simeone
PY - 2026
DA - 2026/08/28
PB - arXiv
DO - 10.48550/arxiv.2608.28179
UR - https://arxiv.org/abs/2608.28179
AB - We study risk-averse decision making, in which an agent selects actions while being uncertain about the true system state. The risk is measured via optimized certainty equivalent (OCE) metrics, which generalize popular criteria such as mean-variance risk and conditional value-at-risk (CVaR). We characterize the optimal policy under known distributions, and show that it reduces to a prediction set-based solution for the CVaR. This provides an operational interpretation of conformal prediction-type prediction sets. For unknown distributions, we develop a data-driven calibration strategy, based on a synthetic model for the likelihood and held-out calibration data, yielding high-probability control of the OCE risk. The approach is evaluated on two wireless beamforming settings.
ER -