This project is a paper review for the course Theoretical Guidelines for High Dimentional Data Analysis by Christophe GIRAUD.
We review the paper False Discoveries Occur Early on the Lasso Path by Weijie Su, Malgorzata Bogdan and Emmanuel Candes (https://arxiv.org/abs/1511.01957).
In our report, we explore the results of this paper to try to find their weaknesses and their limits, first numerically and then theoretically.