Title :
Improvements to the SMO algorithm for SVM regression
Author :
Shevade, S.K. ; Keerthi, S.S. ; Bhattacharyya, C. ; Murthy, K.R.K.
Author_Institution :
Dept. of Comput. Sci. & Autom., Indian Inst. of Sci., Bangalore, India
fDate :
9/1/2000 12:00:00 AM
Abstract :
This paper points out an important source of inefficiency in Smola and Scholkopf´s (1998) sequential minimal optimization (SMO) algorithm for support vector machine regression that is caused by the use of a single threshold value. Using clues from the Karush-Kuhn-Tucker conditions for the dual problem, two threshold parameters are employed to derive modifications of SMO for regression. These modified algorithms perform significantly faster than the original SMO on the datasets tried
Keywords :
neural nets; quadratic programming; statistical analysis; Karush-Kuhn-Tucker conditions; quadratic programming; regression; sequential minimal optimization; support vector machine; threshold parameters; Algorithm design and analysis; Computer science; Iterative algorithms; Kernel; Mathematical programming; Neural networks; Pattern recognition; Quadratic programming; Support vector machine classification; Support vector machines;
Journal_Title :
Neural Networks, IEEE Transactions on
