Semi-supervised Triply Robust Inductive Transfer Learning
share
In this work, we propose a semi-supervised triply robust inductive transfer learning (STRIFLE) approach, which integrates heterogeneous data from label rich source population and label scarce target population to improve the learning accuracy in the target population. Specifically, we consider a high dimensional covariate shift setting and employ two nuisance models, a density ratio model and an imputation model, to combine transfer learning and surrogate-assisted semi-supervised learning strategies organically and achieve triple robustness. While the STRIFLE approach requires the target and source populations to share the same conditional distribution of outcome Y given both the surrogate features S and predictors X, it allows the true underlying model of Y|X to differ between the two populations due to the potential covariate shift in S and X. Different from double robustness, even if both nuisance models are misspecified or the distribution of Y|S,X is not the same between the two populations, when the transferred source population and the target population share enough similarities, the triply robust STRIFLE estimator can still partially utilize the source population, and it is guaranteed to be no worse than the target-only surrogate-assisted semi-supervised estimator with negligible errors. These desirable properties of our estimator are established theoretically and verified in finite-sample via extensive simulation studies. We utilize the STRIFLE estimator to train a Type II diabetes polygenic risk prediction model for the African American target population by transferring knowledge from electronic health records linked genomic data observed in a larger European source population.
READ FULL TEXT