Saved in:
Bibliographic Details
Main Authors: Liu, Jialei, Liao, Jun, Fang, Kuangnan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.10919
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914157831389184
author Liu, Jialei
Liao, Jun
Fang, Kuangnan
author_facet Liu, Jialei
Liao, Jun
Fang, Kuangnan
contents Positive-Unlabeled (PU) learning presents unique challenges due to the lack of explicitly labeled negative samples, particularly in high-stakes domains such as fraud detection and medical diagnosis. To address data scarcity and privacy constraints, we propose a novel transfer learning with model averaging framework that integrates information from heterogeneous data sources - including fully binary labeled, semi-supervised, and PU data sets - without direct data sharing. For each source domain type, a tailored logistic regression model is conducted, and knowledge is transferred to the PU target domain through model averaging. Optimal weights for combining source models are determined via a cross-validation criterion that minimizes the Kullback-Leibler divergence. We establish theoretical guarantees for weight optimality and convergence, covering both misspecified and correctly specified target models, with further extensions to high-dimensional settings using sparsity-penalized estimators. Extensive simulations and real-world credit risk data analyses demonstrate that our method outperforms other comparative methods in terms of predictive accuracy and robustness, especially under limited labeled data and heterogeneous environments.
format Preprint
id arxiv_https___arxiv_org_abs_2511_10919
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Heterogeneous Multisource Transfer Learning via Model Averaging for Positive-Unlabeled Data
Liu, Jialei
Liao, Jun
Fang, Kuangnan
Machine Learning
Positive-Unlabeled (PU) learning presents unique challenges due to the lack of explicitly labeled negative samples, particularly in high-stakes domains such as fraud detection and medical diagnosis. To address data scarcity and privacy constraints, we propose a novel transfer learning with model averaging framework that integrates information from heterogeneous data sources - including fully binary labeled, semi-supervised, and PU data sets - without direct data sharing. For each source domain type, a tailored logistic regression model is conducted, and knowledge is transferred to the PU target domain through model averaging. Optimal weights for combining source models are determined via a cross-validation criterion that minimizes the Kullback-Leibler divergence. We establish theoretical guarantees for weight optimality and convergence, covering both misspecified and correctly specified target models, with further extensions to high-dimensional settings using sparsity-penalized estimators. Extensive simulations and real-world credit risk data analyses demonstrate that our method outperforms other comparative methods in terms of predictive accuracy and robustness, especially under limited labeled data and heterogeneous environments.
title Heterogeneous Multisource Transfer Learning via Model Averaging for Positive-Unlabeled Data
topic Machine Learning
url https://arxiv.org/abs/2511.10919