Consistency Analysis of Regularization of Coefficient Based on Weak Correlation Sampling
International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064


Downloads: 112 | Views: 333

Research Paper | Statistics | China | Volume 6 Issue 9, September 2017 | Popularity: 6.1 / 10


     

Consistency Analysis of Regularization of Coefficient Based on Weak Correlation Sampling

Lu Luo, Xinxin Chang


Abstract: Aiming at the weakly correlated sampling satisfying the strong mixing condition, and the coefficient satisfies the polynomial decay, with the use of the sample operator and the integral operator. The proof of the least squares coefficient regularization algorithm is obtained, and it is concluded that the regularization conditionfor the learning speed of. At the same time, the saturation index of the regularization algorithm based on weak correlation sampling is 2, which shows that the coefficient regularization algorithm has some advantages in learning the smooth function compared with the usual least squares Tikhonov regularization algorithm.


Keywords: Coefficient Regularization Regression Strong Mixing Condition Integral Operator


Edition: Volume 6 Issue 9, September 2017


Pages: 1113 - 1116



Please Disable the Pop-Up Blocker of Web Browser

Verification Code will appear in 2 Seconds ... Wait



Text copied to Clipboard!
Lu Luo, Xinxin Chang, "Consistency Analysis of Regularization of Coefficient Based on Weak Correlation Sampling", International Journal of Science and Research (IJSR), Volume 6 Issue 9, September 2017, pp. 1113-1116, https://www.ijsr.net/getabstract.php?paperid=ART20176771, DOI: https://www.doi.org/10.21275/ART20176771

Top