ComparisonofSVMRegressionwithLeastSquareMethod.docVIP

NC 2569 Under Review in Neural Computation, Revised June 10, 2002 Selecting the Loss Function for Robust Linear Regression Vladimir Cherkassky Yunqian Ma Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, U.S.A. {cherkass, myq}@ece.umn.edu Abstract This paper addresses selection of the loss function for regression problems with finite data. It is well-known (under standard regression formulation) that for a known noise density there exist an optimal loss function under an asymptotic setting (large number of samples), i.e. squared loss is op