Adaptive Double Sparse Iterative Hard Thresholding

Solving high-dimensional double sparse linear regression via an iterative hard thresholding algorithm. Furthermore, the method is extended to jointly estimate multiple graphical models. For more details, please see < https://www.jmlr.org/papers/v25/23-0653.html> and .

ADSIHT

ADSIHT aims to solve the high-dimensional double sparse linear regression problem, that is, a combination of element-wise and group-wise sparsity. In our paper, we show that our algorithm achieves the minimax optimality with fast linear convergence.

Installation

Installation of the ADSIHT package can be done directly from GitHub:

install.packages("remotes")
remotes::install_github("ADSIHT/ADSIHT")

References

Zhang Yanhang, Zhifan Li, and Jianxin Yin. "A minimax optimal approach to high-dimensional double sparse linear regression." arXiv preprint arXiv:2305.04182 (2023).

Contact

Please direct questions and comments to the issues page.