Abstract
The restricted eigenvalue (RE) condition characterizes the sample complexity of accurate recovery in the context of high-dimensional estimators such as Lasso and Dantzig selector (Bickel et al., 2009). Recent work has shown that random design matrices drawn from any thin-tailed (sub- Gaussian) distributions satisfy the RE condition with high probability, when the number of samples scale as the square of the Gaussian width of the restricted set (Banerjee et al., 2014; Tropp, 2015). We pose the equivalent question for heavy-tailed distributions: Given a random design matrix drawn from a heavy-tailed distribution satisfying the small-ball property (Mendelson, 2015), does the design matrix satisfy the RE condition with the same order of sample complexity as sub- Gaussian distributions? An answer to the question will guide the design of high-dimensional estimators for heavy tailed problems.
Original language | English (US) |
---|---|
Journal | Journal of Machine Learning Research |
Volume | 40 |
Issue number | 2015 |
State | Published - 2015 |
Event | 28th Conference on Learning Theory, COLT 2015 - Paris, France Duration: Jul 2 2015 → Jul 6 2015 |
Bibliographical note
Publisher Copyright:© 2015 A. Agarwal & S. Agarwal.