Mathematica: A flexible design environment for neural networks

被引：0
作者：
Duncan, SC
Tweney, RD
机构：
[1] Bowling Green State University, Bowling Green, OH
[2] Department of Psychology, Bowling Green State University, Bowling Green
来源：
BEHAVIOR RESEARCH METHODS INSTRUMENTS & COMPUTERS | 1997年 / 29卷 / 02期
关键词：
Neural Network; Stimulus Onset Asynchrony; Behavior Research Method; Output Activation; Network Code;
D O I：
10.3758/BF03204810
中图分类号：
B841 [心理学研究方法];
学科分类号：
040201 ;
摘要：
Several neural networks were developed in Mathematica in order to explore the role of ''spiky'' neurons in neural network memory simulations. Using Mathematica for this task confirmed its value as a powerful tool for neural network development: It exhibited distinct advantages over other environments in programming ease, flexibility of data structures, and the graphical assessment of network performance.
引用
收藏
页码：194 / 199
页数：6
相关论文
共 11 条
[1]  
DUNCAN S, 1996, 37 ANN M PSYCH SOC C
[2]  
Freeman J.A., 1994, Simulating neural networks with Mathematica
[3]   IMPLEMENTATION OF GLOBAL MEMORY MODELS WITH SOFTWARE THAT DOES SYMBOLIC COMPUTATION [J].
GRONLUND, SD ;
SHEU, CF ;
RATCLIFF, R .
BEHAVIOR RESEARCH METHODS INSTRUMENTS & COMPUTERS, 1990, 22 (02) :228-235
[4]  
Hinton G. E., 1986, PARALLEL DISTRIBUTED, V1, P77
[5]  
KIM C, 1995, P 17 ANN C COGN SCI, P472
[6]   EVENT-RELATED POTENTIAL ANALYSIS USING MATHEMATICA [J].
LORIG, TS ;
URBACH, TP .
BEHAVIOR RESEARCH METHODS INSTRUMENTS & COMPUTERS, 1995, 27 (03) :358-366
[7]  
Masters T., 1993, PRACTICAL NEURAL NET
[8]   USING COMPUTERS IN EMPIRICAL AND THEORETICAL WORK IN COGNITIVE-PSYCHOLOGY [J].
RATCLIFF, R .
BEHAVIOR RESEARCH METHODS INSTRUMENTS & COMPUTERS, 1994, 26 (02) :94-106
[9]  
Rumelhart D. E., 1986, PARALLEL DISTRIBUTED, V1
[10]  
WOLFRAM S, 1996, MATHEMATICA VERSION
12