AI Paper Atlas

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

JMLR 2026 · Label-efficient and robust learning › Semi-supervised learning · approach: Self-training theory

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 changedWhereMeasuredEffect
initialize SIM training with the supervised estimator (Step I) instead of a naive guess (ablation)SASH · Simulation, estimation of gammal2 estimation error of gamma: around 55% higher error without Step I under weak surrogacyImproved
use surrogates via SASH instead of supervised learning only (SL)labeled-data estimator · Simulation, weak surrogatel2 estimation error of beta: 1.834 -> 0.949Improved
use surrogates via SASH instead of SLlabeled-data estimator · Simulation, strong surrogatel2 estimation error of beta: 1.834 -> 0.661Improved
use SASH single-index fitting instead of SS-uLasso (5%)semi-supervised estimator · Simulation, weak surrogatel2 estimation error of beta: 1.038 -> 0.949Improved
use SASH instead of pLassosemi-supervised estimator · Simulation, weak surrogatel2 estimation error of beta: 1.560 -> 0.949Improved
use SASH instead of SS-RMRCE (oracle)semi-supervised estimator · Simulation, strong surrogatel2 estimation error of beta: 1.122 -> 0.661Improved
use privacy-preserving summary-statistic aggregation instead of pooled individual dataSASH vs IPD · Simulation, weak surrogatel2 estimation error of beta: 0.735 -> 0.949Hurt
use stronger surrogates instead of weakSASH · Simulationl2 estimation error of beta: 0.949 -> 0.661Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy