Conjugate Gradient Iterative Hard Thresholding

New 2014

The development of this software and the performance analysis of NIHT, HTP, and CSMPSP has led to the development of a new algorithm which exploits the computational advantages of these algorithms. The CGIHT family of algorithms is introduced in the following paper along with theoretical, uniform recovery guarantees and extensive empirical testing.

CGIHT: Conjugate Gradient Iterative Hard Thresholding for compressed sensing and matrix completion
J.D. Blanchard, J. Tanner, K. Wei
Submitted, April 2014.
DowNload
The performance of the algorithms under the model y=Ax+e for a misfit or noise vector e is provided in the following paper.

Conjugate Gradient Iterative Hard Thresholding: observed noise stability in compressed sensing
J.D. Blanchard, J. Tanner, K. Wei
Submitted, May 2014.