[2604.05446] MEC: Machine-Learning-Assisted Generalized Entropy Calibration for Semi-Supervised Mean Estimation

Statistics > Machine Learning arXiv:2604.05446 (stat) [Submitted on 7 Apr 2026] Title:MEC: Machine-Learning-Assisted Generalized Entropy Calibration for Semi-Supervised Mean Estimation Authors:Se Yoon Lee, Jae Kwang Kim View a PDF of the paper titled MEC: Machine-Learning-Assisted Generalized Entropy Calibration for Semi-Supervised Mean Estimation, by Se Yoon Lee and 1 other authors View PDF HTML (experimental) Abstract:Obtaining high-quality labels is costly, whereas unlabeled covariates are often abundant, motivating semi-supervised inference methods with reliable uncertainty quantification. Prediction-powered inference (PPI) leverages a machine-learning predictor trained on a small labeled sample to improve efficiency, but it can lose efficiency under model misspecification and suffer from coverage distortions due to label reuse. We introduce Machine-Learning-Assisted Generalized Entropy Calibration (MEC), a cross-fitted, calibration-weighted variant of PPI. MEC improves efficiency by reweighting labeled samples to better align with the target population, using a principled calibration framework based on Bregman projections. This yields robustness to affine transformations of the predictor and relaxes requirements for validity by replacing conditions on raw prediction error with weaker projection-error conditions. As a result, MEC attains the semiparametric efficiency bound under weaker assumptions than existing PPI variants. Across simulations and a real-data applicati...