Reproduce “Censored Quantile Regression Neural Networks for Distribution-Free Survival Analysis” 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
Tim Pearce, Jong-Hyeon Jeong, Yichen Jia, Jun Zhu
CQRNN predicts a selected grid of conditional quantiles from right-censored data using one neural network. It estimates each censored observation’s nearest predicted percentile and uses that estimate to weight losses at the censoring bound and a pseudo-outcome. CQRNN had the lowest CensDCal on 10 of 12 real-target datasets, but this calibration result does not establish universal superiority or direct accuracy.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.