Approximate Leave-One-Out Cross Validation

This repository contains packages which implement the approximate leave-one-out cross validation strategy for common regressors in an efficient way. For reference and specific formula, see:

Approximate Leave-One-Out for High-Dimensional Non-Differentiable Learning Problems: Shuaiwen Wang, Wenda Zhou, Arian Maleki, Haiho Lu, Vahab Mirrokni.

The core code is written in C, and is integrated into different languages using their respective packaging systems.