SMO算法由Microsoft Research的John C. Platt在1998年提出，并成为最快的二次规划优化算法，特别针对线性SVM和数据稀疏时性能更优。关于SMO最好的资料就是他本人写的《Sequential Minimal Optimization A Fast Algorithm for Training Support Vector Machines》了。

Make Electronics Learning Through Discovery, 2nd Edition Platt