Maximum Entropy Approach for Generalized Support Vector Machine Optimization Problems
广义支持向量机优化问题的极大熵方法

This paper presents a new approach to solve generalized support vector machine optimization problems. Based on KKT complementarity condition in optimization theory, unconstrained optimization models for support vector machines are built, and an approximate algorithm-maximum entropy method is given. Theory analysis and primary numerical results illustrate that maximum entropy method for support vector machines is feasible and effective.