Common Subset Selection of Inputs in Multiresponse Regression推荐.pdf

Publication V c 2006 IEEE. Reprinted, with permission, from Timo Simil¨a and Jarkko Tikka (2006). Common subset selection of in- puts in multiresponse regression, Proceedings of the IEEE International Joint Conference on Neural Networks - IJCNN 2006, pp. 1908–1915. This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of Helsinki University of Technology’s products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or pro- motional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@. By choosing to view this document, you agree to all provisions of the copyright laws protecting it. V 2006 International Joint Conference on Neural Networks Sheraton Vancouver Wall Centre Hotel, Vancouver, BC, Canada July 16-21, 2006 Common Subset Selection of Inputs in Multiresponse Regression ¨ Timo Simila and Jarkko Tikka Abstract— We propose the Multiresponse Sparse Regression recognize important combinations of inputs, especially when algorithm, an input selection method for the purpose of esti- collinearities are present [2]. Subset selection is a hard mating several response variables. It is a forward selection pro- combinatorial problem in general. On the contrary, subset cedure for linearly parameterized models, which updates with carefully chosen step lengths. The step length rule extends the selection coupled with shrinkag