Research map / Learning theory and algorithms / Prediction and decision losses
Empirical Risk Minimization under Random Censorship
Guillaume Ausset, Stephan Clémençon, François Portier
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 |
|---|---|---|---|
| add IPCW weighting instead of naive risk (ablation) | Linear Regression · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 4.971 -> 3.193 | Improved |
| add IPCW weighting instead of observed-only risk (ablation) | Linear Regression · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 3.898 -> 3.193 | Improved |
| add IPCW weighting instead of naive risk (ablation) | Ridge · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 4.962 -> 3.193 | Improved |
| add IPCW weighting instead of naive risk (ablation) | Random Forest · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 2.636 -> 2.577 | Improved |
| add IPCW weighting instead of naive risk (ablation) | Kernel Ridge · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 2.704 -> 2.683 | Improved |
| add IPCW weighting instead of naive risk (ablation) | SVR · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 2.796 -> 2.768 | Improved |
| add IPCW weighting instead of naive risk (ablation) | Random Forest · TCGA Cancer (RNA, 18% censoring) | Harrell concordance: 0.603 -> 0.630 | Improved |
| add IPCW weighting instead of naive risk (ablation) | SVR · TCGA Cancer (RNA, 18% censoring) | Harrell concordance: 0.575 -> 0.563 | Hurt |
| replace Cox model by Random Forest with IPCW (ablation) | Cox · TCGA Cancer (RNA, 18% censoring) | L2 error (years): 18.78 -> 2.577 | Improved |
| replace Cox model by Random Forest with IPCW (ablation) | Cox · TCGA Cancer (RNA, 18% censoring) | Harrell concordance: 0.6095 -> 0.630 | Improved |
| replace Random Survival Forest by Ridge + IPCW (p=0.25) (ablation) | Random Survival Forest · Cox simulated data (p=0.25, n=200) | prediction error: 2.78 -> 1.24 | Improved |
| replace Cox Proportional Hazards by Ridge + IPCW (p=0.25) (ablation) | Cox Proportional Hazards · Cox simulated data (p=0.25, n=200) | prediction error: 7.03 -> 1.24 | Improved |
Closest papers by meaning
- Boosting Methods for Interval-censored Data with Regression and Classification (ICLR 2025 · similarity 0.85)
- Risk Minimization from Adaptively Collected Data: Guarantees for Supervised and Policy Learning (NeurIPS 2021 · similarity 0.84)
- Statistical Robustness of Empirical Risks in Machine Learning (JMLR 2023 · similarity 0.84)
- On the Asymptotic Distribution of the Minimum Empirical Risk (ICML 2024 · similarity 0.83)
- Scaling-up Empirical Risk Minimization: Optimization of Incomplete $U$-statistics (JMLR 2016 · similarity 0.83)
- Learning Survival Models with Right-Censored Reporting Delays (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.
