Error Analysis of Generalized Nystrom Kernel Regression

被引:0
|
作者
Chen, Hong [1 ]
Xia, Haifeng [2 ]
Cai, Weidong [3 ]
Huang, Heng [1 ]
机构
[1] Univ Texas Arlington, Comp Sci & Engn, Arlington, TX 76019 USA
[2] Huazhong Agr Univ, Math & Stat, Wuhan 430070, Peoples R China
[3] Univ Sydney, Sch Informat Technol, Sydney, NSW 2006, Australia
来源
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 29 (NIPS 2016) | 2016年 / 29卷
基金
中国国家自然科学基金; 美国国家科学基金会;
关键词
APPROXIMATION; MATRIX;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Nystrom method has been successfully used to improve the computational efficiency of kernel ridge regression (KRR). Recently, theoretical analysis of Nystrom KRR, including generalization bound and convergence rate, has been established based on reproducing kernel Hilbert space (RKHS) associated with the symmetric positive semi-definite kernel. However, in real world applications, RKHS is not always optimal and kernel function is not necessary to be symmetric or positive semi-definite. In this paper, we consider the generalized Nystrom kernel regression (GNKR) with l(2) coefficient regularization, where the kernel just requires the continuity and boundedness. Error analysis is provided to characterize its generalization performance and the column norm sampling strategy is introduced to construct the refined hypothesis space. In particular, the fast learning rate with polynomial decay is reached for the GNKR. Experimental analysis demonstrates the satisfactory performance of GNKR with the column norm sampling.
引用
收藏
页数:9
相关论文
共 50 条
  • [31] Nystrom Approximation for Sparse Kernel Methods: Theoretical Analysis and Empirical Evaluation
    Xu, Zenglin
    Jin, Rong
    Shen, Bin
    Zhu, Shenghuo
    PROCEEDINGS OF THE TWENTY-NINTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2015, : 3115 - 3121
  • [32] Efficient Nystrom method for Low Rank Approximation and Error Analysis
    Patel, Lokendra Singh
    Saha, Suman
    Ghrera, S. P.
    2015 THIRD INTERNATIONAL CONFERENCE ON IMAGE INFORMATION PROCESSING (ICIIP), 2015, : 536 - 542
  • [33] Analysis of regularized Nystrom subsampling for regression functions of low smoothness
    Lu, Shuai
    Mathe, Peter
    Pereverzyev, Sergiy, Jr.
    ANALYSIS AND APPLICATIONS, 2019, 17 (06) : 931 - 946
  • [34] Hashing with Generalized Nystrom Approximation
    Yun, Jeong-Min
    Kim, Saehoon
    Choi, Seungjin
    12TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2012), 2012, : 1188 - 1193
  • [35] MEAN SQUARED ERROR PROPERTIES OF KERNEL ESTIMATES OF REGRESSION QUANTILES
    JONES, MC
    HALL, P
    STATISTICS & PROBABILITY LETTERS, 1990, 10 (04) : 283 - 289
  • [36] Classification of Error-Diffused Halftone Images Based on Spectral Regression Kernel Discriminant Analysis
    Zeng, Zhigao
    Wen, Zhiqiang
    Yi, Shengqiu
    Zeng, Sanyou
    Zhu, Yanhui
    Liu, Qiang
    Tong, Qi
    ADVANCES IN MULTIMEDIA, 2016, 2016 (2016)
  • [37] The effect of the regularity of the error process on the performance of kernel regression estimators
    Karim Benhenni
    Mustapha Rachdi
    Yingcai Su
    Metrika, 2013, 76 : 765 - 781
  • [38] The effect of the regularity of the error process on the performance of kernel regression estimators
    Benhenni, Karim
    Rachdi, Mustapha
    Su, Yingcai
    METRIKA, 2013, 76 (06) : 765 - 781
  • [39] From Polynomial Fitting to Kernel Ridge Regression: A Generalized Difference Filter for Encoder Signal Analysis
    Zhao, Ming
    Ma, Zhipeng
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2020, 69 (09) : 6212 - 6220
  • [40] An approach for constructing parsimonious generalized Gaussian kernel regression models
    Wang, XX
    Chen, S
    Brown, DJ
    NEUROCOMPUTING, 2004, 62 (1-4) : 441 - 457