Improvements to Platt's SMO Algorithm for SVM Classifier Design
Top Cited Papers
- 1 March 2001
- journal article
- Published by MIT Press in Neural Computation
- Vol. 13 (3) , 637-649
- https://doi.org/10.1162/089976601300014493
Abstract
This article points out an important source of inefficiency in Platt's sequential minimal optimization (SMO) algorithm that is caused by the use of a single threshold value. Using clues from the KKT conditions for the dual problem, two threshold parameters are employed to derive modifications of SMO. These modified algorithms perform significantly faster than the original SMO on all benchmark data sets tried.Keywords
This publication has 1 reference indexed in Scilit:
- Robust linear programming discrimination of two linearly inseparable setsOptimization Methods and Software, 1992