基于神经网络的回归学习算法收敛性分析
기우신경망락적회귀학습산법수렴성분석
Convergence Analysis of Regression Learning Algorithm by Neural Networks
  • 万方数据
    工程数学学报 공정수학학보
    CHINESE JOURNAL OF ENGINEERING MATHEMATICS
    2012年 4期 493-498 ,共6页

    张永全%曹飞龙%戴腾辉

    장영전%조비룡%대등휘

    回归%神经网络%覆盖数%收敛率 迴歸%神經網絡%覆蓋數%收斂率
    회귀%신경망락%복개수%수렴솔
    regression%neural Network%covering number%convergence rate
    本文利用最小二乘理论研究学习理论中的回归问题.其目的在于利用概率不等式与神经网络的逼近性质来分析回归学习算法的误差.结论表明,当回归函数满足一定的光滑性时,得到较为紧的上界且该上界与输入空间的维数无关.
    본문이용최소이승이론연구학습이론중적회귀문제.기목적재우이용개솔불등식여신경망락적핍근성질래분석회귀학습산법적오차.결론표명,당회귀함수만족일정적광활성시,득도교위긴적상계차해상계여수입공간적유수무관.
    This article considers the convergence rate of regression learning algorithm via the approximation property of neural networks and covering number.The upper bounds of convergence rate provided by our results is considerably tight and independent of the dimension of input space when the target function satisfies certain smooth conditions.Hence the curse of dimensionality is alleviated in these cases.
    원문보기 원문다운로드 사이트로이동
저자의 최근 논문