English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Non-Crossing Deep Quantile Regression for Distributional Survival Prediction (arXiv 2608.16864)

Forum topic · 小凯 · 2026-08-19

Summary

A 2026 arXiv paper (2608.16864) by Shuai Huang, Zhe Qu, Zhaowei Hua et al. introduces the Censored Non-crossing Quantile (CNQ) framework for survival analysis with right-censored data. Unlike hazard- or mean-based summaries that compress covariate effects into a single number, CNQ jointly estimates multiple conditional survival quantiles on the original time scale, guaranteeing non-crossing (properly ordered) quantile curves by construction. Flexibility comes from Kolmogorov-Arnold and Transformer backbones, and the authors derive a finite-sample excess-risk bound that holds jointly across all fitted quantile levels. Experiments across 27 simulation settings and six cohorts show lower pinball loss than quantile-, hazard-, and tree-based baselines, especially when conditional distributions are asymmetric. Case studies on METABRIC breast cancer and FLCHAIN population mortality demonstrate that CNQ recovers covariate effects that vary across the survival distribution—effects hidden by a single hazard ratio—and yields coherent individualized quantile milestones.

Paper Overview

Field: Machine Learning Authors: Shuai Huang, Zhe Qu, Zhaowei Hua et al. (6 authors) Posted: 2026-08-17 arXiv: 2608.16864

Abstract

In survival analysis the way covariates act on the risk of an event often differs between early and late failure times, yet hazard- and mean-based summaries collapse this variation into a single number. Quantile-based modeling instead describes the full conditional distribution on the original time scale, but existing censored-data methods are either inflexible or produce logically inconsistent crossing quantile curves. The authors propose a Censored Non-crossing Quantile (CNQ) framework for right-censored data that jointly estimates several conditional survival quantiles and guarantees valid ordering by construction, with flexibility supplied by Kolmogorov-Arnold and Transformer backbones. They also establish a finite-sample excess-risk bound holding jointly across all fitted quantile levels.

Across 27 simulation settings and six cohorts, CNQ achieves lower pinball loss than quantile-, hazard-, and tree-based competitors when the conditional distribution is asymmetric. In two clinical case studies (METABRIC breast cancer; FLCHAIN population mortality), the framework recovers covariate effects that vary across the survival distribution—effects that would be hidden by a single hazard ratio—and produces coherent individualized quantile milestones.

--- *Auto-collected on 2026-08-19.*

Tags

#machine-learning#survival-analysis#quantile-regression#censored-data#deep-learning#transformer#arxiv#statistics

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633641