Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.06531 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908357454987264 |
|---|---|
| author | Cao, Yong-Syun Imori, Shinpei Ing, Ching-Kang |
| author_facet | Cao, Yong-Syun Imori, Shinpei Ing, Ching-Kang |
| contents | Imori and Ing (2025) proposed the importance-weighted orthogonal greedy algorithm (IWOGA) for model selection in high-dimensional misspecified regression models under covariate shift. To determine the number of IWOGA iterations, they introduced the high-dimensional importance-weighted information criterion (HDIWIC). They argued that the combined use of IWOGA and HDIWIC, IWOGA + HDIWIC, achieves an optimal trade-off between variance and squared bias, leading to optimal convergence rates in terms of conditional mean squared prediction error. In this article, we provide a theoretical justification for this claim by establishing the optimality of IWOGA + HDIWIC under a set of reasonable assumptions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_06531 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | High-Dimensional Importance-Weighted Information Criteria: Theory and Optimality Cao, Yong-Syun Imori, Shinpei Ing, Ching-Kang Machine Learning Statistics Theory Imori and Ing (2025) proposed the importance-weighted orthogonal greedy algorithm (IWOGA) for model selection in high-dimensional misspecified regression models under covariate shift. To determine the number of IWOGA iterations, they introduced the high-dimensional importance-weighted information criterion (HDIWIC). They argued that the combined use of IWOGA and HDIWIC, IWOGA + HDIWIC, achieves an optimal trade-off between variance and squared bias, leading to optimal convergence rates in terms of conditional mean squared prediction error. In this article, we provide a theoretical justification for this claim by establishing the optimality of IWOGA + HDIWIC under a set of reasonable assumptions. |
| title | High-Dimensional Importance-Weighted Information Criteria: Theory and Optimality |
| topic | Machine Learning Statistics Theory |
| url | https://arxiv.org/abs/2505.06531 |