<ul><li>The random feature model in high-dimensional regression has gained attention in machine learning literature.</li><li>This paper studies the performance of the random features model for non-iid feature vectors.</li><li>The study utilizes the notion of variance profile from random matrix theory.</li><li>The paper provides asymptotic equivalents for the risks of ridge regression with random features in a high-dimensional framework.</li></ul>

High-dimensional ridge regression with random features for non-identically distributed data with a variance profile

Discover more