Reproduce “Rectifying Conformity Scores for Better Conditional Coverage” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Vincent Plassier, Alexander Fishkov, Victor Dheur, Mohsen Guizani, Souhaib Ben Taieb, Maxim Panov, Eric Moulines
RCP adapts multi-output prediction regions by estimating an input-conditional quantile of a univariate conformity score, rectifying scores, and applying split conformal calibration. This retains marginal coverage under the method’s conditions while aiming to improve conditional coverage without estimating the full response distribution. On households, its WSC was 0.905 versus 0.825 for PCP; this is dataset-specific, not universal.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.