Research map / Learning theory and algorithms / Prediction and decision losses
Support Vector Hazards Machine: A Counting Process Framework for Learning Risk Scores for Censored Outcomes
Yuanjia Wang, Tianle Chen, Donglin Zeng
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 SVHM instead of modified SVR | support vector learning for censored data · Sim AFT censoring (60% censored, n=100) | RMSE: 6.00 -> 5.18 | Improved |
| use SVHM instead of IPCW-KM | support vector learning for censored data · Sim AFT censoring (60% censored, n=100) | RMSE: 6.45 -> 5.18 | Improved |
| estimate censoring distribution with Cox model instead of Kaplan-Meier in IPCW (ablation) | IPCW · Sim AFT censoring (60% censored, n=100) | RMSE: 6.45 -> 6.56 | No clear effect |
| increase sample size from 100 to 200 (ablation) | SVHM · Sim AFT censoring (40% censored, 0 noise) | RMSE: 4.68 -> 4.49 | Improved |
| use Gaussian kernel instead of linear kernel (ablation) | SVHM · Sim AFT censoring (60% censored, n=100) | RMSE: 5.18 -> 5.26 | No clear effect |
| use SVHM instead of Cox model when data follow AFT | Cox model + 1-NN · Sim AFT-generated (10 noise vars, n=100) | RMSE: 0.766 -> 0.739 | Improved |
| use SVHM instead of Cox model when data follow Cox model | Cox model + 1-NN · Sim Cox-generated (10 noise vars, n=100) | RMSE: 6.068 -> 6.099 | No clear effect |
| predict with 1-NN on fitted scores instead of Cox median survival time (ablation) | Cox model · Sim AFT-generated (10 noise vars, n=100) | RMSE: 0.950 -> 0.766 | Improved |
| use 1-NN instead of 3-NN prediction (ablation) | SVHM · Sim Cox-generated (10 noise vars, n=100) | RMSE: 6.503 -> 6.099 | Improved |
| use SVHM linear instead of IPCW-KM | risk score learning · PREDICT-HD (without MRI) | C-index: 0.47 -> 0.73 | Improved |
| add structural MRI imaging biomarkers (ablation) | SVHM (Linear) · PREDICT-HD | C-index: 0.73 -> 0.74 | Improved |
| use Gaussian kernel instead of linear kernel (ablation) | SVHM · PREDICT-HD (with MRI) | C-index: 0.74 -> 0.79 | Improved |
Closest papers by meaning
- Multitask Boosting for Survival Analysis with Competing Risks (NeurIPS 2018 · similarity 0.86)
- Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data (NeurIPS 2025 · similarity 0.85)
- Boosting Methods for Interval-censored Data with Regression and Classification (ICLR 2025 · similarity 0.84)
- Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring (NeurIPS 2026 · similarity 0.83)
- History-Aware Conformal Prediction Sets for Censored Time-to-Event Outcomes (NeurIPS 2026 · similarity 0.83)
- Deep Copula-Based Survival Analysis for Dependent Censoring with Identifiability Guarantees (AAAI 2024 · similarity 0.83)
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.
