有效的v支持向量回归机的v解路径算法

摘要

由Sch(o)lkopf 等人提出的v支持向量回归机具有通过参数v控制支持向量和错误向量个数的优点,然而与标准的支持向量机相比,其形式更为复杂,迄今为止仍没有有效的算法计算v解路径.基于v支持向量回归机的修改形式,提出了一种新的解路径算法,它能够追踪参数v对应的所有解,并通过理论分析和实验,说明了该算法能够尽可能地避免不可行的更新路径,并在有限步内拟合出所有的v解路径.%The v-support vector regression (v-SVR) proposed by Scholkopf, et al., has the advantage of using the parameter v to control the number of support vectors and margin errors, however, compared to ￡-SVR, its formulation is more complicated. Until now, there have been no effective methods used to compute the v-path for it. This paper proposes a new solution path algorithm, which is designed based on a modified formulation of v-SVR and traces the solution path with respect to the parameter v. Through theoretical analysis and experiments, results can show that the algorithm can avoid the infeasible updating path, and fit the entire v-path in finite steps.