Research map / Probabilistic and scientific modeling / Variational and generative sampling
Online Learning and Inference for Cox Proportional Hazards Model Using Renewable Sieve Estimation
Mengtong Hu, Ziyang Gong, Xu Shi, Ling Zhou, Peter Song
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Bernstein sieve renewable updates instead of multi-epoch offline SGD | COLSA · Simulated Cox streams (K=200) | computation time (s): 58.14 -> 28.36 | Improved |
| use sieve basis expansion with numerical integration rather than Meta/Online updates | COLSA · Simulated Cox streams (K=200) | computation time (s): 4.516 -> 28.36 | Hurt |
| use COLSA instead of Online piecewise-constant baseline when batch size is small (nk = 100) | Online survival method · Simulated Cox streams (nk=100) | bias vs Oracle | Improved |
| vary pre-estimation factor rho over 1.5, 2.0, 2.5, 3.0 (ablation) | COLSA · Simulated Cox streams | estimation performance | No clear effect |
| use COLSA instead of Meta on TCGA gene inference | Meta · TCGA pan-cancer | significance decisions matching Oracle: Meta missed 5 Oracle-significant genes | Improved |
| use COLSA instead of Online and SGD on TCGA gene inference | Online and SGD · TCGA pan-cancer | decisions disagreeing with Oracle: 12 and 10 disagreements | Improved |
| use COLSA instead of SGD on SRTR sparse categorical covariates | SGD · SRTR kidney transplant | Z-statistic, Donor Obesity: 9.58 -> 1.74 (Oracle 1.63) | Improved |
| use COLSA instead of Online on Hispanic recipient effect | COLSA · SRTR kidney transplant | inference agreement with Oracle: -2.22 -> -1.58 | No clear effect |
Closest papers by meaning
- Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data (NeurIPS 2025 · similarity 0.86)
- Functional Martingale Residual Process for High-Dimensional Cox Regression with Model Averaging (JMLR 2020 · similarity 0.85)
- Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring (NeurIPS 2026 · similarity 0.84)
- FastSurvival: Hidden Computational Blessings in Training Cox Proportional Hazards Models (NeurIPS 2024 · similarity 0.83)
- Longitudinal Targeted Minimum Loss-based Estimation with Temporal-Difference Heterogeneous Transformer (ICML 2024 · similarity 0.83)
- SurvivalPFN: Amortizing Survival Prediction via In-Context Bayesian Inference (NeurIPS 2026 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
