Efficient techniques for simultaneous variable selection and sensor selection via convex selection inducing penalties

Show full item record

Title: Efficient techniques for simultaneous variable selection and sensor selection via convex selection inducing penalties
Author: Fokoue, Ernest
Abstract: This paper extends results from the traditional D-optimality machinery to derive an efficient technique for simultaneous variable selection and sensor selection. An important advantage of the proposed technique is the convexity of the formulated optimization task along with a byproduct of straightforward sparsity. The theoretical foundation of the proposed method is explored at great length, and a variety of examples are provided to demonstrated the effectiveness of our technique. Comparisons with existing techniques are offered that provide evidence as the superiority of our technique on a variety of indicators.
Record URI: http://hdl.handle.net/1850/13086
Date: 2011-01-03

Files in this item

Files Size Format View
EFokoueArticle01-03-2011.pdf 162.6Kb PDF View/Open

The following license files are associated with this item:

This item appears in the following Collection(s)

Show full item record

Search RIT DML


Advanced Search

Browse