AI RESEARCH
Distribution-Aware Conformal Prediction: A Framework for generating efficient prediction intervals for time series
arXiv CS.LG
•
ArXi:2605.26569v1 Announce Type: new We present Distribution-aware Conformal Prediction (DCP), a unified framework integrating probabilistic predictors like Monte Carlo dropout, deep ensembles, and quantile regression with score-agnostic conformal calibration to produce valid and efficient prediction intervals. Leveraging a numerical inversion approach to construct interval bounds, DCP accommodates arbitrary combinations of distribution generating predictors and nonconformity scores.