Using genetic algorithms to select architecture of a feedforward artificial neural network

被引:120
作者
Arifovic, J
Gençay, R
机构
[1] Univ Windsor, Dept Econ, Windsor, ON N9B 3P4, Canada
[2] Simon Fraser Univ, Dept Econ, Burnaby, BC V5A 1N6, Canada
[3] Bilkent Univ, Dept Econ, TR-06533 Ankara, Turkey
来源
PHYSICA A | 2001年 / 289卷 / 3-4期
关键词
genetic algorithms; neural networks; model selection;
D O I
10.1016/S0378-4371(00)00479-9
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
This paper proposes a model selection methodology for feedfoward network models based on the genetic algorithms and makes a number of distinct but inter-related contributions to the model selection literature for the feedfoward networks. First, we construct a genetic algorithm which can search for the global optimum of an arbitrary function as the output of a feedforward network model. Second, we allow the genetic algorithm to evolve the type of inputs, the number of hidden units and the connection structure between the inputs and the output layers. Third, we study how introduction of a local elitist procedure which we call the election operator affects the algorithm's performance. We conduct a Monte Carlo simulation to study the sensitiveness of the global approximation properties of the studied genetic algorithm. Finally, we apply the proposed methodology to the daily foreign exchange returns. (C) 2001 Published by Elsevier Science B.V. All rights reserved.
引用
收藏
页码:574 / 594
页数:21
相关论文
共 41 条
[1]   GENETIC ALGORITHM LEARNING AND THE COBWEB MODEL [J].
ARIFOVIC, J .
JOURNAL OF ECONOMIC DYNAMICS & CONTROL, 1994, 18 (01) :3-28
[2]  
COHEN JE, 1986, RANDOM MATRICES THEI, V50
[3]  
Cybenko G., 1989, Mathematics of Control, Signals, and Systems, V2, P303, DOI 10.1007/BF02551274
[4]   The topological invariance of Lyapunov exponents in embedded dynamics [J].
Dechert, WD ;
Gencay, R .
PHYSICA D, 1996, 90 (1-2) :40-55
[5]  
DECHERT WD, 1992, J APPL EC, V7, P41
[6]   FINDING STRUCTURE IN TIME [J].
ELMAN, JL .
COGNITIVE SCIENCE, 1990, 14 (02) :179-211
[7]   EVOLVING NEURAL NETWORKS [J].
FOGEL, DB ;
FOGEL, LJ ;
PORTO, VW .
BIOLOGICAL CYBERNETICS, 1990, 63 (06) :487-493
[8]   ON THE APPROXIMATE REALIZATION OF CONTINUOUS-MAPPINGS BY NEURAL NETWORKS [J].
FUNAHASHI, K .
NEURAL NETWORKS, 1989, 2 (03) :183-192
[9]   ON LEARNING THE DERIVATIVES OF AN UNKNOWN MAPPING WITH MULTILAYER FEEDFORWARD NETWORKS [J].
GALLANT, AR ;
WHITE, H .
NEURAL NETWORKS, 1992, 5 (01) :129-138
[10]  
GALLANT AR, 1998, P 2 ANN IEEE C NEUR, pI657