Newton-Raphson Meets Sparsity: Sparse Learning Via a Novel Penalty and a Fast Solver

被引:4
|
作者
Cao, Yongxiu [1 ]
Kang, Lican [2 ]
Li, Xuerui [2 ]
Liu, Yanyan [2 ]
Luo, Yuan [2 ]
Shi, Yueyong [3 ,4 ]
机构
[1] Zhongnan Univ Econ & Law, Sch Stat & Math, Wuhan 430073, Peoples R China
[2] Wuhan Univ, Sch Math & Stat, Wuhan 430072, Peoples R China
[3] China Univ Geosci, Sch Econ & Management, Wuhan 430074, Peoples R China
[4] China Univ Geosci, Ctr Resources & Environm Econ Res, Wuhan 430074, Peoples R China
基金
中国国家自然科学基金;
关键词
Interpolation; Smoothing methods; Linear regression; Computational modeling; Optimization; Numerical models; Learning systems; Cubic Hermite interpolation penalty (CHIP); high-dimensional linear regression; Karush-Kuhn-Tucker (KKT); nonasymptotic error bound; support detection-based Newton-Raphson (SDNR); NONCONVEX PENALIZED REGRESSION; GENERALIZED LINEAR-MODELS; DUAL ACTIVE SET; VARIABLE SELECTION; ALGORITHM; REGULARIZATION;
D O I
10.1109/TNNLS.2023.3251748
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In machine learning and statistics, the penalized regression methods are the main tools for variable selection (or feature selection) in high-dimensional sparse data analysis. Due to the nonsmoothness of the associated thresholding operators of commonly used penalties such as the least absolute shrinkage and selection operator (LASSO), the smoothly clipped absolute deviation (SCAD), and the minimax concave penalty (MCP), the classical Newton-Raphson algorithm cannot be used. In this article, we propose a cubic Hermite interpolation penalty (CHIP) with a smoothing thresholding operator. Theoretically, we establish the nonasymptotic estimation error bounds for the global minimizer of the CHIP penalized high-dimensional linear regression. Moreover, we show that the estimated support coincides with the target support with a high probability. We derive the Karush-Kuhn-Tucker (KKT) condition for the CHIP penalized estimator and then develop a support detection-based Newton-Raphson (SDNR) algorithm to solve it. Simulation studies demonstrate that the proposed method performs well in a wide range of finite sample situations. We also illustrate the application of our method with a real data example.
引用
收藏
页码:12057 / 12067
页数:11
相关论文
共 50 条
  • [1] A Fast Newton-Raphson Method in stochastic linearization
    Canor, Thomas
    Blaise, Nicolas
    Denoel, Vincent
    EURODYN 2014: IX INTERNATIONAL CONFERENCE ON STRUCTURAL DYNAMICS, 2014, : 2839 - 2844
  • [2] Fast Newton-Raphson Power Flow Analysis Based on Sparse Techniques and Parallel Processing
    Ahmadi, Afshin
    Smith, Melissa C.
    Collins, Edward R.
    Dargahi, Vahid
    Jin, Shuangshuang
    IEEE TRANSACTIONS ON POWER SYSTEMS, 2022, 37 (03) : 1695 - 1705
  • [3] COVARIANCE FACTORIZATION VIA NEWTON-RAPHSON ITERATION
    ANDERSON, BDO
    IEEE TRANSACTIONS ON INFORMATION THEORY, 1978, 24 (02) : 183 - 187
  • [4] A novel method for calculation of Newton-Raphson electric load flow based on sparse matrix
    Kamel, Salah
    Abdel-Akher, Mamdouh
    Jurado, Francisco
    DYNA, 2017, 92 (04): : 376 - 376
  • [5] Fast parallel Newton-Raphson power flow solver for large number of system calculations with CPU and GPU
    Wang, Zhenqi
    Wende-von Berg, Sebastian
    Braun, Martin
    SUSTAINABLE ENERGY GRIDS & NETWORKS, 2021, 27 (27):
  • [6] Floating-point photonic iterative solver demonstrated for Newton-Raphson method
    Klein, Andrew B.
    Zhu, Zheyuan
    Li, Guifang
    Pang, Shuo
    APPLIED PHYSICS LETTERS, 2024, 125 (12)
  • [7] Newton-Raphson Solver for Finite Element Methods Featuring Nonlinear Hysteresis Models
    Chama, Abdoulkadri
    Gerber, Stiaan
    Wang, Rong-Jie
    IEEE TRANSACTIONS ON MAGNETICS, 2018, 54 (01)
  • [8] Newton-Raphson Nonlinear Solver for Electric Field Integral Equation and Resistive Boundary Condition
    Prakash, Jay
    Greenaway, Mark T.
    Cools, Kristof
    2024 IEEE INTERNATIONAL SYMPOSIUM ON ANTENNAS AND PROPAGATION AND INC/USNCURSI RADIO SCIENCE MEETING, AP-S/INC-USNC-URSI 2024, 2024, : 483 - 484
  • [9] Transversality enforced Newton-Raphson algorithm for fast calculation of maximum loadability
    Ali, Mazhar
    Dymarsky, Anatoly
    Turitsyn, Konstantin
    IET GENERATION TRANSMISSION & DISTRIBUTION, 2018, 12 (08) : 1729 - 1737
  • [10] Fast Parameter Analysis of the Complex Exponential Signal by the Newton-Raphson Method
    Lin, Whei-Min
    Su, Tzu-Jung
    Wu, Rong-Ching
    Tsai, Jong-Ian
    ICIEA: 2009 4TH IEEE CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS, VOLS 1-6, 2009, : 3872 - +