高级检索
    朱烨雷 王玉军 罗 强 陶 卿. 一种求解截断Hinge损失的软阈值坐标下降算法[J]. 计算机研究与发展, 2013, 50(11): 2295-2303.
    引用本文: 朱烨雷 王玉军 罗 强 陶 卿. 一种求解截断Hinge损失的软阈值坐标下降算法[J]. 计算机研究与发展, 2013, 50(11): 2295-2303.
    Zhu Yelei, Wang Yujun, Luo Qiang, and Tao Qing. A Soft-Thresholding Coordinate Descent Algorithm for Solving Truncated Hinge Loss[J]. Journal of Computer Research and Development, 2013, 50(11): 2295-2303.
    Citation: Zhu Yelei, Wang Yujun, Luo Qiang, and Tao Qing. A Soft-Thresholding Coordinate Descent Algorithm for Solving Truncated Hinge Loss[J]. Journal of Computer Research and Development, 2013, 50(11): 2295-2303.

    一种求解截断Hinge损失的软阈值坐标下降算法

    A Soft-Thresholding Coordinate Descent Algorithm for Solving Truncated Hinge Loss

    • 摘要: 有效地减少支持向量数目能够提高分类器的鲁棒性和精确性,缩短支持向量机(support vector machine, SVM)的训练和测试时间.在众多稀疏算法中,截断Hinge损失方法可以显著降低支持向量的数目,但却导致了非凸优化问题.一些研究者使用CCCP(concave-convex procedure)方法将非凸问题转化为多阶段凸问题求解,不仅增加了额外计算量,而且只能得到局部最优解.为了弥补上述不足,提出了一种基于CCCP的软阈值坐标下降算法.用坐标下降方法求解CCCP子阶段凸问题,提高计算效率;对偶SVM中引入软阈值投影技巧,能够减少更多的支持向量数目,同时选择合适的正则化参数可消除局部最优解的不良影响,提高分类器的分类精度.仿真数据库、UCI数据库和大规模真实数据库的实验证实了所提算法正确性和有效性.

       

      Abstract: Substantially lessening the number of support vectors (SV) can improve robustness to outliers and deliver more accurate classifiers, and reduce the training and predicting time of SVM. Among various sparse algorithms, the truncated Hinge loss method can reduce the number of support vectors effectively. However, the formulated optimization problem is non-convex which can not be solved by conventional convex optimization method. Therefore, some researchers change the non-convex problem into a series of convex problems with concave-convex procedure (CCCP) method, not only increasing the additional amount of calculation but also falling into a local optimal solution. In order to circumvent these drawbacks, we propose a CCCP-based soft-thresholding(ST) coordinate descent (CD) algorithm (STCD-CCCP). We solve CCCP sub-stage convex problem with coordinate descent algorithm to improve computational efficiency. Moreover, we use soft-thresholding projection technique in dual SVM problem to reduce bigger number of support vectors as a result of the algorithm setting the coefficient of the classifier to zero. Meanwhile, soft-thresholding projection technique can eliminate the adverse effect of the local optimal solution and improve classification accuracy by selecting appropriate regularization parameter. Extensive experiments on simulation database, UCI database and large-scale real database have verified that the proposed algorithm is correct and effective.

       

    /

    返回文章
    返回