On Convergence Rates of Mixtures of Polynomial Experts

被引：7
|
作者：
Mendes, Eduardo F. [1 ]
Jiang, Wenxin [1 ]
机构：
[1] Northwestern Univ, Dept Stat, Evanston, IL 60208 USA
来源：
NEURAL COMPUTATION | 2012年 / 24卷 / 11期
关键词：
GENERALIZED LINEAR-MODELS; HIERARCHICAL MIXTURES; OF-EXPERTS; TIME-SERIES; BAYESIAN-INFERENCE; LOCAL MIXTURES; APPROXIMATION; REGRESSION; ESTIMATORS; SIEVE;
D O I：
10.1162/NECO_a_00354
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this letter, we consider a mixture-of-experts structure where m experts are mixed, with each expert being related to a polynomial regression model of order k. We study the convergence rate of the maximum likelihood estimator in terms of how fast the Hellinger distance of the estimated density converges to the true density, when the sample size n increases. The convergence rate is found to be dependent on both m and k, while certain choices of m and k are found to produce near-optimal convergence rates.
引用
收藏
页码：3025 / 3051
页数：27
相关论文
共 50 条
  • [21] OPTIMAL RATES OF CONVERGENCE FOR THE PIECEWISE POLYNOMIAL ESTIMATOR WITH THE INDEX CHOSEN BY THE FPE SELECTION RULE
    CHEN, KW
    COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1988, 17 (09) : 2887 - 2906
  • [22] RATES OF CONVERGENCE FOR THE POSTERIOR DISTRIBUTIONS OF MIXTURES OF BETAS AND ADAPTIVE NONPARAMETRIC ESTIMATION OF THE DENSITY
    Rousseau, Judith
    ANNALS OF STATISTICS, 2010, 38 (01): : 146 - 180
  • [23] Entropies and rates of convergence for maximum likelihood and Bayes estimation for mixtures of normal densities
    Ghosal, S
    Van der Vaart, AW
    ANNALS OF STATISTICS, 2001, 29 (05): : 1233 - 1263
  • [24] Adaptive Mixtures of Local Experts
    Jacobs, Robert A.
    Jordan, Michael I.
    Nowlan, Steven J.
    Hinton, Geoffrey E.
    NEURAL COMPUTATION, 1991, 3 (01) : 79 - 87
  • [25] On the identifiability of mixtures-of-experts
    Jiang, W
    Tanner, MA
    NEURAL NETWORKS, 1999, 12 (09) : 1253 - 1258
  • [26] Functional mixtures-of-experts
    Chamroukhi, Faicel
    Pham, Nhat Thien
    Hoang, Van Ha
    Mclachlan, Geoffrey J.
    STATISTICS AND COMPUTING, 2024, 34 (03)
  • [27] Bayesian methods for mixtures of experts
    Waterhouse, S
    MacKay, D
    Robinson, T
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 8: PROCEEDINGS OF THE 1995 CONFERENCE, 1996, 8 : 351 - 357
  • [28] On the identifiability of mixtures-of-experts
    Department of Statistics, Northwestern University, Evanston, IL 60208, United States
    Neural Netw., 9 (1253-1258):
  • [29] Simple Proof of Dynkin's Formula for Single-Server Systems and Polynomial Convergence Rates
    Veretennikov, A. Yu.
    Zverkina, G. A.
    MARKOV PROCESSES AND RELATED FIELDS, 2014, 20 (03) : 479 - 504
  • [30] The convergence of orthogonal polynomial series
    Kolmogorov, A
    COMPTES RENDUS DE L ACADEMIE DES SCIENCES DE L URSS, 1934, 2 : 291 - 294
12345