tag:blogger.com,1999:blog-6141980.post1718632235159462795..comments2009-05-29T10:14:41.134-05:00Comments on Nuit Blanche: CS: Topics in Compressed Sensing, GraDes and other...Igornoreply@blogger.comBlogger2125tag:blogger.com,1999:blog-6141980.post-548171032663268842009-05-29T10:14:41.134-05:002009-05-29T10:14:41.134-05:00Sorry, got it. In the way they select the update f...Sorry, got it. In the way they select the update factor gamma: tuned to the RIP constant. Right? IHT set this constant to 1 (Authors mention however possible extension to other value).JackDhttp://www.blogger.com/profile/14728376685283207988noreply@blogger.comtag:blogger.com,1999:blog-6141980.post-28608936733179941402009-05-29T09:22:40.534-05:002009-05-29T09:22:40.534-05:00About the GraDes method proposed in "Gradient Desc...About the GraDes method proposed in "Gradient Descent with Sparsification: An iterative algorithm for sparse recovery with restricted isometry property", I cannot see the difference between this method and the Iterative Hard Thresholding (IHT) by T. Blumensath and M. Davies. <br /><br />If anybody could help me to understand where the difference is ...<br /><br />JackJackDhttp://www.blogger.com/profile/14728376685283207988noreply@blogger.com