English

Transfer Learning for Nonparametric Regression: Non-asymptotic Minimax Analysis and Adaptive Procedure

Machine Learning 2024-01-24 v1 Machine Learning

Abstract

Transfer learning for nonparametric regression is considered. We first study the non-asymptotic minimax risk for this problem and develop a novel estimator called the confidence thresholding estimator, which is shown to achieve the minimax optimal risk up to a logarithmic factor. Our results demonstrate two unique phenomena in transfer learning: auto-smoothing and super-acceleration, which differentiate it from nonparametric regression in a traditional setting. We then propose a data-driven algorithm that adaptively achieves the minimax risk up to a logarithmic factor across a wide range of parameter spaces. Simulation studies are conducted to evaluate the numerical performance of the adaptive transfer learning algorithm, and a real-world example is provided to demonstrate the benefits of the proposed method.

Keywords

Cite

@article{arxiv.2401.12272,
  title  = {Transfer Learning for Nonparametric Regression: Non-asymptotic Minimax Analysis and Adaptive Procedure},
  author = {T. Tony Cai and Hongming Pu},
  journal= {arXiv preprint arXiv:2401.12272},
  year   = {2024}
}
R2 v1 2026-06-28T14:23:59.178Z