tag:blogger.com,1999:blog-6555947.post112965657717147057..comments2014-01-12T10:46:48.153-07:00Comments on The Geomblog: Algorithms and Technologies, and ESA 2006Suresh Venkatasubramanianhttps://plus.google.com/112165457714968997350noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-6555947.post-1130279480903941222005-10-25T16:31:00.000-06:002005-10-25T16:31:00.000-06:00"...it is not often that one can invoke a sophisti..."<I>...it is not often that one can invoke a sophisticated algorithm that has known worst-case bounds, and find an implementation for it.</I>"<BR/><BR/>I find it intriguing that the sophisticated implementations (at least for linear programming) don't actually achieve the worst-case bounds. For efficiency in practice they make parameter choices which result in non-polynomial worst-case bounds (or at least, in versions for which a polynomial bound isn't known). Of course, you can always combine the two by running for a long time with the "efficient" parameter choices, and then if you haven't terminated yet, switching to the worst-case parameter choices.Anonymousnoreply@blogger.com