OPTIMIZATION FOR TRAINING NEURAL NETS

被引：122
作者：
BARNARD, E
机构：
[1] Department of Electronics and Computer Engineering, University of Pretoria, 0002, Pretoria
来源：
IEEE TRANSACTIONS ON NEURAL NETWORKS | 1992年 / 3卷 / 02期
关键词：
D O I：
10.1109/72.125864
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Various techniques of optimizing criterion functions to train neural-net classifiers are investigated. These techniques include three standard deterministic techniques (variable metric, conjugate gradient, and steepest descent), and a new stochastic technique. It is found that the stochastic technique is preferable on problems with large training sets and that the convergence rates of the variable-metric and conjugate-gradient techniques are similar.
引用
收藏
页码：232 / 240
页数：9
相关论文
共 26 条
  • [1] [Anonymous], 1987, LEARNING INTERNAL RE
  • [2] A COMPARISON BETWEEN CRITERION FUNCTIONS FOR LINEAR CLASSIFIERS, WITH AN APPLICATION TO NEURAL NETS
    BARNARD, E
    CASASENT, D
    [J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1989, 19 (05): : 1030 - 1041
  • [3] BARNARD E, 1989, JUN P INT JOINT C NE, P111
  • [4] EXPERIMENTS ON NEURAL NET RECOGNITION OF SPOKEN AND WRITTEN TEXT
    BURR, DJ
    [J]. IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1988, 36 (07): : 1162 - 1168
  • [5] CATER JP, 1987, JUL P C NEUR NETW SA, P645
  • [6] DENNIS JE, 1983, NUMERICAL METHODS UN
  • [7] Duda R. O., 1973, PATTERN CLASSIFICATI, V3
  • [8] FAHLMAN SE, 1988, 1988 P CONN MOD SUMM, P38
  • [9] Gill P. E., 1981, PRACTICAL OPTIMIZATI
  • [10] ANALYSIS OF HIDDEN UNITS IN A LAYERED NETWORK TRAINED TO CLASSIFY SONAR TARGETS
    GORMAN, RP
    SEJNOWSKI, TJ
    [J]. NEURAL NETWORKS, 1988, 1 (01) : 75 - 89
123