EUVIMEDDescoperă EUVIMED
Open access

LEARNER: A Transfer Learning Method for Low‐Rank Matrix Estimation

Sean McGrath iD, Cenhao Zhu, Ryan O'Dea, Min Guo, Rui Duan iD

Statistics in Medicine · 2026-07 · 45 (15-17)

DOI: 10.1002/sim.70669 ↗

At a glance

Journal article in Statistics in Medicine, published online on 2026-07-12. Written by Sean McGrath and 4 co-authors. Institution: Department of Biostatistics, Yale School of Public Health, New Haven, Connecticut, USA. Major MEDLINE topics: Transfer Machine Learning, Models, Statistical, Machine Learning. 49 references. The full text is freely available in PubMed Central. Funded by: NIGMS NIH HHS, NIH HHS, National Science Foundation.

RezumatSubjectsDate bibliograficeCiteRelated studies

Abstract (original language)

Low‐rank matrix estimation is a fundamental problem in statistics and machine learning with applications across biomedical sciences, including genetics, medical imaging, drug discovery, and electronic health record data analysis. In the context of heterogeneous data generated from diverse sources, a key challenge lies in leveraging data from a source population to enhance the estimation of a low‐rank matrix in a target population of interest. We propose an approach that leverages similarity in the latent row and column spaces between the source and target populations to improve estimation in the target population, which we refer to as LatEnt spAce‐based tRaNsfer lEaRning (LEARNER). LEARNER is based on performing a low‐rank approximation of the target population data which penalizes differences between the latent row and column spaces between the source and target populations. We present a cross‐validation approach that allows the method to adapt to the degree of heterogeneity across populations. We conducted extensive simulations which found that LEARNER often outperforms the benchmark approach that only uses the target population data, especially as the signal‐to‐noise ratio in the source population increases. We also performed an illustrative application and empirical comparison of LEARNER and benchmark approaches in a re‐analysis of summary statistics from a genome‐wide association study in the BioBank Japan cohort. LEARNER is implemented in the R package learner and the Python package learner‐py .

Subjects

MeSH terms (MEDLINE)

Transfer Machine LearningModels, StatisticalMachine LearningHumansComputer SimulationSignal-To-Noise RatioAlgorithms

Author keywords

genome‐wide association studiesheterogeneous data sourceslatent spaceslow‐rank matrix estimationtransfer learning

Authors and institutions

  • Department of Biostatistics, Yale School of Public Health, New Haven, Connecticut, USA.
  • Operations Research Center, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.
  • Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA.
  • Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA.

ORCID profiles: Sean McGrath, Rui Duan

Funding

  • NIGMS NIH HHS
  • NIH HHS
  • National Science Foundation
  • National Institutes of Health

Cite

APA

McGrath, S., Zhu, C., O'Dea, R., Guo, M., & Duan, R. (2026). LEARNER: A Transfer Learning Method for Low‐Rank Matrix Estimation. Statistics in Medicine, 45(15-17). https://doi.org/10.1002/sim.70669

Vancouver

McGrath S, Zhu C, O'Dea R, Guo M, Duan R. LEARNER: A Transfer Learning Method for Low‐Rank Matrix Estimation. Statistics in Medicine. 2026;45(15-17). doi:10.1002/sim.70669

BibTeX
@article{McGrath2026,
  author = {McGrath, Sean and Zhu, Cenhao and O'Dea, Ryan and Guo, Min and Duan, Rui},
  title = {LEARNER: A Transfer Learning Method for Low‐Rank Matrix Estimation},
  journal = {Statistics in Medicine},
  year = {2026},
  volume = {45},
  number = {15-17},
  doi = {10.1002/sim.70669}
}
Start for free →