Kernel Matrix Approximation for Parameters Tuning of Support Vector Regression

摘要：

Parameters tuning is fundamental for support vector regression (SVR). Previous tuning methods mainly adopted a nested two-layer optimization framework, where the inner one solved a standard SVR for fixed hyper-parameters and the outer one adjusted the hyper-parameters, which directly led to high computational complexity. To solve this problem, we propose a kernel matrix approximation algorithm KMA-a based on Monte Carlo and incomplete Cholesky factorization. The KMA-a approximates a given kernel matrix by a low-rank matrix, which will be used to feed SVR to improve its performance and farther accelerate the whole parameters tuning process. Finally, on the basis of the computational complexity analysis of the KMA-α, we verify/the performance improvement of parameters tuning attributed to the KMA-α on benchmark databases. Theoretical and experimental results show that the KMA-a is a valid and efficient kernel matrix approximation algorithm for parameters tuning of SVR.

作者: Lizhong Ding Shizhong Liao

作者单位: School of Computer Science and Technology, Tianjin University, Tianjin 300072, China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 214-218

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）