Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 53 min ago
Peer-reviewedCC BYJournal of the American Statistical Association

Representation Retrieval Learning for Heterogeneous Data Integration.

  • Qi Xu1
  • Annie Qu2
  1. 1Department of Statistics & Data Science, Carnegie Mellon University, Pittsburgh, PA.
  2. 2Department of Statistics and Applied Probability, University of California, Santa Barbara, Santa Barbara, CA.
Read full textPDFDOICite

Abstract

In the era of big data, large-scale, multi-source, multi-modality datasets are increasingly ubiquitous, offering unprecedented opportunities for predictive modeling and scientific discovery. However, these datasets often exhibit complex heterogeneity, such as covariates shift, posterior drift, and blockwise missingness, which worsen predictive performance of existing supervised learning algorithms. To address these challenges simultaneously, we propose a novel Representation Retrieval ( R2 ) framework, which integrates a dictionary of representation learning modules (representer dictionary) with data source-specific sparsityinduced machine learning model (learners). Under the R2 framework, we introduce the notion of integrativeness for each representer, and propose a novel Selective Integration Penalty (SIP) to explicitly encourage more integrative representers to improve predictive performance. Theoretically, we show that the excess risk bound of the R2 framework is characterized by the integrativeness of representers, and SIP effectively improves the excess risk. Extensive simulation studies validate the superior performance of R2 framework and the effect of SIP. We further apply our method to two real-world datasets to confirm its empirical success. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Related

  1. Diagonal Multi-omics Integration of Heterogeneous Datasets
    Maksim V. Kukushkin, Mikhail S. Arbatskiy, Dmitriy E. Balandin +1
    arXivPreprintCC BY 4.0Full textAnalysis & PDEsStatisticsarXiv
  2. Retrieval-Based In-Context Learning: A Domain Adaptation Framework
    Yilun Zhu, Naihao Deng, Yingcong Li +2
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  3. BRIDGE: Bilevel Retrieval-Credit-Aware Agentic Reinforcement Learning
    Quan Xiao, Mingda Liu, Gaowen Liu +2
    arXivPreprintarXiv licenceFull textOptimisation & controlarXiv
  4. Transfer Learning with Heterogeneous Feature Spaces in Linear Regression
    Zejing Zheng, Rui Huang, Junlong Zhao
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  5. Amortized Probabilistic Retrieval of Atmospheric CO2 from OCO-2 Spectra Using Deep Learning with Laplace Approximations and Normalizing Flows
    Alejandro Calle-Saldarriaga, Felix Jimenez, Jack Grosskreuz +3
    arXivPreprintCC BY 4.0Full textStatisticsarXiv

Cite

BibTeXRIS
BibTeX
@article{xu2026representation,
  title = {{Representation Retrieval Learning for Heterogeneous Data Integration.}},
  author = {Qi Xu and Annie Qu},
  journal = {Journal of the American Statistical Association},
  year = {2026},
  month = sep,
  doi = {10.1080/01621459.2026.2696485},
  url = {https://europepmc.org/article/MED/42846695},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Representation Retrieval Learning for Heterogeneous Data Integration.
AU  - Qi Xu
AU  - Annie Qu
PY  - 2026
DA  - 2026/09/01
JO  - Journal of the American Statistical Association
DO  - 10.1080/01621459.2026.2696485
UR  - https://europepmc.org/article/MED/42846695
AB  - In the era of big data, large-scale, multi-source, multi-modality datasets are increasingly ubiquitous, offering unprecedented opportunities for predictive modeling and scientific discovery. However, these datasets often exhibit complex heterogeneity, such as covariates shift, posterior drift, and blockwise missingness, which worsen predictive performance of existing supervised learning algorithms. To address these challenges simultaneously, we propose a novel Representation Retrieval ( R2 ) framework, which integrates a dictionary of representation learning modules (representer dictionary) with data source-specific sparsityinduced machine learning model (learners). Under the R2 framework, we introduce the notion of integrativeness for each representer, and propose a novel Selective Integration Penalty (SIP) to explicitly encourage more integrative representers to improve predictive performance. Theoretically, we show that the excess risk bound of the R2 framework is characterized by the integrativeness of representers, and SIP effectively improves the excess risk. Extensive simulation studies validate the superior performance of R2 framework and the effect of SIP. We further apply our method to two real-world datasets to confirm its empirical success. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.
SN  - 0162-1459
SN  - 1537-274X
ER  -