基于非独立同分布样本的系数正则化回归算法的误差分析
Error Analysis of Coefficient Regularized Regression and Non-Iid Sampling
DOI: 10.12677/SA.2017.64050, PDF, HTML, XML, 下载: 1,616  浏览: 3,739 
作者: 常欣欣：河北工业大学，天津；陈爽：北京石油化工学院，北京
关键词: 系数正则化回归；不定核；强混合条件；积分算子；Coefficient Regularization Regression； Indefinite Kernel； Strong Mixing Condition； Integral Operator
摘要: 本文分析了基于非独立同分布样本的最小二乘系数正则化算法的误差。全文的框架不同于以往的经典核学习方法。核函数仅仅满足连续性和一致有界性；我们进行基于输出样本满足广义矩理论的误差分析，而不再考虑标准的输出有界假设。最后通过利用积分算子技术得到了满意的与容量无关的误差界和学习率。
Abstract: In this paper, we study the error analysis of indefinite kernel network with coefficient regulariza-tion for non-iid sampling. The framework under investigation is different from classical kernel learning. The kernel function satisfies only the continuity and uniform boundedness; the standard bound assumption for output data is abandoned and we carry out the error analysis with output sample values satisfying a generalized moment hypothesis. Satisfactory error bounds and learning rates independent of capacity are derived by the techniques of integral operator for this learning algorithm.
文章引用：常欣欣, 陈爽. 基于非独立同分布样本的系数正则化回归算法的误差分析[J]. 统计学与应用, 2017, 6(4): 442-454. https://doi.org/10.12677/SA.2017.64050

参考文献

[1] Chun, X.R. and Sun, H.W. (2013) Regularized Least Square Regression with Unbounded and Dependent Sampling. Abstract and Applied Analysis, 2013, 900-914.
[2] Gao, Q. and Ye, P.X. (2016) Coefficient-Based Regularized Regression with Dependent and Unbounded Sampling. International Journal of Wavelets Multiresolution and Information Processing, 14, Article ID: 1650039.
[3] Sun, H. and Guo, Q. (2011) Coefficient Regularized Regression with Non-Iid Sampling. International Journal of Computer Mathematics, 88, 3113-3124.
https://doi.org/10.1080/00207160.2011.587511
[4] Zhang, M.J. and Sun, H.W. (2017) Regression Learning with Non-Identically and Non-Independently Sampling. International Journal of Wavelets Multiresolution and Information Processing, 15, Article ID: 1750007.
https://doi.org/10.1142/S0219691317500072
[5] Vapink, V. (1998) Statistical Learning Theory. Wiley, New York, 3185.
[6] Pan, Z.W. and Xiao, Q.W. (2009) Least-Square Regularized Regression with Non-Iid Sampling. Journal of Statistical Planning and Inference, 139, 3579-3587.
https://doi.org/10.1016/j.jspi.2009.04.007
[7] Sun, H.W. and Wu, Q. (2012) Regu-larized Least Square Regression with Dependent Sampling. Advances in Computational Mathematics, 32, 175-189.
[8] Qiang, W., Ying, Y. and Ding, X.Z. (2006) Learning Rates of Least-Square Regularized Regression. Foundations of Computational Mathematics, 6, 171-192.
https://doi.org/10.1007/s10208-004-0155-9
[9] Sheng, B.H., Ye, P.X. and Wang, J.L. (2012) Learning Rates for Least Square Regression with Coefficient Regularization. Acta Mathmation Sinica, English Series, 8, 2205-2212.
https://doi.org/10.1007/s10114-012-0607-0
[10] Sun, H. and Wu, Q. (2011) Least Square Regression with Indefinite Kernels and Coefficient Regularization. Applied and Computation Harmonic Analysis, 30, 96-109.
https://doi.org/10.1016/j.acha.2010.04.001
[11] Zuo, L., Li, L.Q. and Chen, C. (2015) The Graph Based Semi-Supervised Al-gorithm with l1-Regularizer. Neurocomputing, 149, 966-974.
[12] Qu, Z.F. and Sun, H.W. (2016) Indefinite Kernel Network with lq-Norm Regularization. Discrete Dynamics in Nature and Society Volume, Article ID: 6516258.
[13] Wu, Q. (2013) Regularization Networks with Indefinite Kernels. Journal of Approximation Theory, 166, 1-18.
[14] Cai, J. (2013) Coefficient-Based Regression with Non-Identical Unbounded Sampling. Hindawi Publishing Corporation Abstract and Applied Analysis, 415-425.
https://doi.org/10.1155/2013/134727
[15] Cai, J. and Wang, C (2013) Coefficient-Based Regularized Regression with Indefinite Kernels by Unbounded Sampling. Scientia Sinica Mathematica, 43, 613-624.
[16] Cucker, F. and Zhou, D.X. (2007) Learning Theory: An Approximation Theory Viewpoint. Cambridge University Press.
https://doi.org/10.1017/CBO9780511618796
[17] Sun, H. and Guo, Q. (2009) A Note On Application of Integral Operator in Learning Theory. Applied and Computational Harmonic Analysis, 26, 416-421.

为你推荐