Mathematics - Colloquium
Friday, January 31, 2014
11:00 AM-12:00 PM
ABSTRACT: We propose a new method named calibrated multivariate regression (CMR) for fitting high dimensional multivariate regression models. Compared to existing methods, CMR calibrates the regularization for each regression task with respect to its noise level so that it is simultaneously tuning insensitive and achieves an improved finite sample performance. Theoretically, it is proved that CMR achieves the optimal rate of convergence in parameter estimation. We illustrate the usefulness of CMR by thorough numerical simulations. We also apply CMR on a brain activity prediction problem and find that CMR even outperforms the handcrafted models created by human experts.
Suggested Audiences: College, Adult
Last Modified: January 23, 2014 at 12:04 PM