Multi-step Training of a Generalized Linear Classifier

被引:0
|
作者
Kanishka Tyagi
Michael Manry
机构
[1] The University of Texas at Arlington,Department of Electrical Engineering
来源
Neural Processing Letters | 2019年 / 50卷
关键词
Linear classifiers; Nonlinear functions; Pruning; Orthogonal least squares; Newton’s algorithm;
D O I
暂无
中图分类号
学科分类号
摘要
We propose a multi-step training method for designing generalized linear classifiers. First, an initial multi-class linear classifier is found through regression. Then validation error is minimized by pruning of unnecessary inputs. Simultaneously, desired outputs are improved via a method similar to the Ho-Kashyap rule. Next, the output discriminants are scaled to be net functions of sigmoidal output units in a generalized linear classifier. This classifier is trained via Newton’s algorithm. Performance gains are demonstrated at each step. Using widely available datasets, the final network’s tenfold testing error is shown to be less than that of several other linear and generalized linear classifiers reported in the literature.
引用
收藏
页码:1341 / 1360
页数:19
相关论文
共 50 条
  • [41] Multi-step methods for equations
    Kumar S.
    Sharma J.R.
    Argyros I.K.
    ANNALI DELL'UNIVERSITA' DI FERRARA, 2024, 70 (4) : 1193 - 1215
  • [42] What is a 'step' in a multi-step pathogenesis of leukemia?
    Jankovic, GM
    Colovic, MD
    Bogdanovic, AD
    Vukanic, D
    Andolina, M
    Anagnostopoulos, A
    LEUKEMIA RESEARCH, 1996, 20 (06) : 531 - 532
  • [43] A novelty Multi-Step Associated with Laplace Transform Semi Analytic Technique for Solving Generalized Non-linear Differential Equations
    Al-Jizani, Khalid Hammood
    Abud, Ahmed Hanoon
    BAGHDAD SCIENCE JOURNAL, 2023, 20 (05) : 2265 - 2270
  • [44] Brain tumor grade classification using multi-step pre-training
    Mehmood, Yasar
    Bajwa, Usama Ijaz
    Anwar, Muhammad Waqas
    INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY, 2024, 34 (02)
  • [45] A MULTI-STEP APPROACHTO BUILDING AND TESTING A VACCINE MISINFORMATION TRAINING FOR HEALTHCARE PROFESSIONALS
    Michel, Alexandra
    Schulz, Gretchen
    Limaye, Rupali
    ANNALS OF BEHAVIORAL MEDICINE, 2023, 57 : S484 - S484
  • [46] DEEP RESIDUAL NETWORKS WITH COMMON LINEAR MULTI-STEP AND ADVANCED NUMERICAL SCHEMES
    Luo, Zhengbo
    Zhou, Weilian
    Kamata, Sei-ichiro
    Hu, Xuehui
    2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, 2022, : 3286 - 3290
  • [47] On greedy multi-step inertial randomized Kaczmarz method for solving linear systems
    Su, Yansheng
    Han, Deren
    Zeng, Yun
    Xie, Jiaxin
    CALCOLO, 2024, 61 (04)
  • [48] Linear multi-step methods and their numerical stability for solving gradient flow equations
    Qiong-Ao Huang
    Wei Jiang
    Jerry Zhijian Yang
    Gengen Zhang
    Advances in Computational Mathematics, 2023, 49
  • [49] Multi-step linear prediction based speech dereverberation in noisy reverberant environment
    Kinoshita, Keisuke
    Delcroix, Marc
    Nakatani, Tomohiro
    Miyoshi, Masato
    INTERSPEECH 2007: 8TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION, VOLS 1-4, 2007, : 1085 - 1088
  • [50] Multi-step tangential versus transversal linearizations in non-linear dynamics
    Ghosh, S.
    Roy, D.
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2007, 72 (05) : 582 - 605