In the problem of classification （or pattern recognition）, given a set of n samples, we attempt to construct a classifier gn with a small misclassification error. It is important to study the convergence rates of th...In the problem of classification （or pattern recognition）, given a set of n samples, we attempt to construct a classifier gn with a small misclassification error. It is important to study the convergence rates of the misclassification error as n tends to infinity. It is known that such a rate can＇t exist for the set of all distributions. In this paper we obtain the optimal convergence rates for a class of distributions L^（λ,ω） in multicategory classification and nonstandard binary classification.展开更多
Recently,Li[16]introduced three kinds of single-hidden layer feed-forward neural networks with optimized piecewise linear activation functions and fixed weights,and obtained the upper and lower bound estimations on th...Recently,Li[16]introduced three kinds of single-hidden layer feed-forward neural networks with optimized piecewise linear activation functions and fixed weights,and obtained the upper and lower bound estimations on the approximation accuracy of the FNNs,for continuous function defined on bounded intervals.In the present paper,we point out that there are some errors both in the definitions of the FNNs and in the proof of the upper estimations in[16].By using new methods,we also give right approximation rate estimations of the approximation by Li’s neural networks.展开更多
Let B^pΩ,1≤Р≤∞,be the set of all bounded functions in L^p（R）which can be extended to entire unctions of exponential typeΩ. The unitbrm bounds for truncation error of Shannon sampling expansion fromlocal averag...Let B^pΩ,1≤Р≤∞,be the set of all bounded functions in L^p（R）which can be extended to entire unctions of exponential typeΩ. The unitbrm bounds for truncation error of Shannon sampling expansion fromlocal averages are obtained for functions f∈BpΩwith the decay condition f（t）≤A/t^δ,t≠0,where A and δare positive constants. Furthermore we also establish similar results for non-bandlimit functions in Besov classes with the same decay condition as above.展开更多
基金Research supported in part by NSF of China under Grants 10571010 and 10171007The work was partially done while the first author was visiting the Institute for Mathematical Sciences, National University of Singapore in 2003The visit was supported by the Institute
文摘In the problem of classification （or pattern recognition）, given a set of n samples, we attempt to construct a classifier gn with a small misclassification error. It is important to study the convergence rates of the misclassification error as n tends to infinity. It is known that such a rate can＇t exist for the set of all distributions. In this paper we obtain the optimal convergence rates for a class of distributions L^（λ,ω） in multicategory classification and nonstandard binary classification.
文摘Recently,Li[16]introduced three kinds of single-hidden layer feed-forward neural networks with optimized piecewise linear activation functions and fixed weights,and obtained the upper and lower bound estimations on the approximation accuracy of the FNNs,for continuous function defined on bounded intervals.In the present paper,we point out that there are some errors both in the definitions of the FNNs and in the proof of the upper estimations in[16].By using new methods,we also give right approximation rate estimations of the approximation by Li’s neural networks.
基金Supported by the National Natural Science Foundation of China(Nos.61379014 and 11271199)
文摘Let B^pΩ,1≤Р≤∞,be the set of all bounded functions in L^p（R）which can be extended to entire unctions of exponential typeΩ. The unitbrm bounds for truncation error of Shannon sampling expansion fromlocal averages are obtained for functions f∈BpΩwith the decay condition f（t）≤A/t^δ,t≠0,where A and δare positive constants. Furthermore we also establish similar results for non-bandlimit functions in Besov classes with the same decay condition as above.
