Research map / Label-efficient and robust learning / Semi-supervised learning
Efficient Modeling of Surrogates to Improve Multi-source High-dimensional Integrative Regression
Yue Liu, Molei Liu, Zijian Guo, Tianxi Cai
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 |
|---|---|---|---|
| initialize SIM training with the supervised estimator (Step I) instead of a naive guess (ablation) | SASH · Simulation, estimation of gamma | l2 estimation error of gamma: around 55% higher error without Step I under weak surrogacy | Improved |
| use surrogates via SASH instead of supervised learning only (SL) | labeled-data estimator · Simulation, weak surrogate | l2 estimation error of beta: 1.834 -> 0.949 | Improved |
| use surrogates via SASH instead of SL | labeled-data estimator · Simulation, strong surrogate | l2 estimation error of beta: 1.834 -> 0.661 | Improved |
| use SASH single-index fitting instead of SS-uLasso (5%) | semi-supervised estimator · Simulation, weak surrogate | l2 estimation error of beta: 1.038 -> 0.949 | Improved |
| use SASH instead of pLasso | semi-supervised estimator · Simulation, weak surrogate | l2 estimation error of beta: 1.560 -> 0.949 | Improved |
| use SASH instead of SS-RMRCE (oracle) | semi-supervised estimator · Simulation, strong surrogate | l2 estimation error of beta: 1.122 -> 0.661 | Improved |
| use privacy-preserving summary-statistic aggregation instead of pooled individual data | SASH vs IPD · Simulation, weak surrogate | l2 estimation error of beta: 0.735 -> 0.949 | Hurt |
| use stronger surrogates instead of weak | SASH · Simulation | l2 estimation error of beta: 0.949 -> 0.661 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PASS — PASS: Part-Aware Self-Supervised Pre-training for Person Re-identification (ECCV 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Surrogate Assisted Semi-supervised Inference for High Dimensional Risk Prediction (JMLR 2023 · similarity 0.90)
- High-dimensional Analysis of Knowledge Distillation: Weak-to-Strong Generalization and Scaling Laws (ICLR 2025 · similarity 0.86)
- Continuous Treatment Effects with Surrogate Outcomes (ICML 2024 · similarity 0.83)
- Scaling laws for learning with real and surrogate data (NeurIPS 2024 · similarity 0.83)
- Bridging Fairness and Efficiency in Conformal Inference: A Surrogate-Assisted Group-Clustered Approach (ICML 2025 · similarity 0.82)
- Towards the Efficient Inference by Incorporating Automated Computational Phenotypes under Covariate Shift (ICML 2025 · 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.
