Based on the statistical learning theory and optimization theory , svms have been successfully applied to many fields such as pattern recognition , regression and etc . training an svm amounts to solving a convex quadratic programming problem . in this paper we do some researches on svms by the optimization theory and method 它將機器學習問題轉(zhuǎn)化為求解最優(yōu)化問題,并應(yīng)用最優(yōu)化理論構(gòu)造算法來解決問題,本文主要是從最優(yōu)化理論和算法的角度對支持向量機中的最優(yōu)化問題進行研究。