Convergence analysis of online algorithms

被引:21
|
作者
Ying, Yiming [1 ]
机构
[1] City Univ Hong Kong, Dept Math, Kowloon, Hong Kong, Peoples R China
关键词
online learning algorithm; reproducing kernel Hilbert space; regularized sample error; general loss function;
D O I
10.1007/s10444-005-9002-z
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
In this paper, we are interested in the analysis of regularized online algorithms associated with reproducing kernel Hilbert spaces. General conditions on the loss function and step sizes are given to ensure convergence. Explicit learning rates are also given for particular step sizes.
引用
收藏
页码:273 / 291
页数:19
相关论文
共 50 条
  • [1] Convergence analysis of online algorithms
    Yiming Ying
    Advances in Computational Mathematics, 2007, 27 : 273 - 291
  • [2] Convergence of Unregularized Online Learning Algorithms
    Lei, Yunwen
    Shi, Lei
    Guo, Zheng-Chu
    JOURNAL OF MACHINE LEARNING RESEARCH, 2018, 18
  • [4] Fast and strong convergence of online learning algorithms
    Zheng-Chu Guo
    Lei Shi
    Advances in Computational Mathematics, 2019, 45 : 2745 - 2770
  • [5] Fast Convergence of Online Pairwise Learning Algorithms
    Boissier, Martin
    Lyu, Siwei
    Ying, Yiming
    Zhou, Ding-Xuan
    ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 51, 2016, 51 : 204 - 212
  • [6] Fast and strong convergence of online learning algorithms
    Guo, Zheng-Chu
    Shi, Lei
    ADVANCES IN COMPUTATIONAL MATHEMATICS, 2019, 45 (5-6) : 2745 - 2770
  • [7] QUICKEST CONVERGENCE OF ONLINE ALGORITHMS VIA DATA SELECTION
    Romero, Daniel
    Berberidis, Dimitris
    Giannakis, Georgios B.
    2016 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING PROCEEDINGS, 2016, : 6185 - 6189
  • [8] Convergence analysis of stochastic algorithms
    Shapiro, A
    Wardi, Y
    MATHEMATICS OF OPERATIONS RESEARCH, 1996, 21 (03) : 615 - 628
  • [9] Formal Analysis of Online Algorithms
    Aminof, Benjamin
    Kupferman, Orna
    Lampert, Robby
    AUTOMATED TECHNOLOGY FOR VERIFICATION AND ANALYSIS, 2011, 6996 : 213 - +
  • [10] CONVERGENCE ANALYSIS OF CANONICAL GENETIC ALGORITHMS
    RUDOLPH, G
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (01): : 96 - 101