A higher order Bayesian neural network with spiking units

被引:48
作者
Lansner, A [1 ]
Holst, A [1 ]
机构
[1] ROYAL INST TECHNOL,DEPT NUMER ANAL & COMP SCI,S-10044 STOCKHOLM,SWEDEN
关键词
D O I
10.1142/S0129065796000816
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We treat a Bayesian confidence propagation neural network, primarily in a classifier context. The one-layer version of the network implements a naive Bayesian classifier, which requires the input attributes to be independent. This limitation is overcome by a higher order network. The higher order Bayesian neural network is evaluated on a real world task of diagnosing a telephone exchange computer. By introducing stochastic spiking units, and soft interval coding, it is also possible to handle uncertain as well as continuous valued inputs.
引用
收藏
页码:115 / 128
页数:14
相关论文
共 54 条
[41]  
Namatame A., 1992, International Journal of Neural Systems, V3, P19, DOI 10.1142/S0129065792000036
[42]  
Pearl, 1989, MORGAN KAUFMANN SERI, DOI DOI 10.1016/C2009-0-27609-4
[43]  
PROTZEL PW, 1991, ARTIFICIAL NEURAL NETWORKS, VOLS 1 AND 2, P273
[44]   HIGHER-ORDER ASSOCIATIVE MEMORIES AND THEIR OPTICAL IMPLEMENTATIONS [J].
PSALTIS, D ;
PARK, CH ;
HONG, J .
NEURAL NETWORKS, 1988, 1 (02) :149-163
[45]   ITERATED PATTERNS OF BRAIN CIRCUITRY (OR HOW THE CORTEX GETS ITS SPOTS) [J].
PURVES, D ;
RIDDLE, DR ;
LAMANTIA, AS .
TRENDS IN NEUROSCIENCES, 1992, 15 (10) :362-368
[46]  
REDDING NJ, 1991, ARTIFICIAL NEURAL NETWORKS, VOLS 1 AND 2, P25
[47]   REDUNDANCY REDUCTION AS A STRATEGY FOR UNSUPERVISED LEARNING [J].
REDLICH, AN .
NEURAL COMPUTATION, 1993, 5 (02) :289-304
[48]  
Salinas E, 1994, J Comput Neurosci, V1, P89, DOI 10.1007/BF00962720
[49]   LEARNING FACTORIAL CODES BY PREDICTABILITY MINIMIZATION [J].
SCHMIDHUBER, J .
NEURAL COMPUTATION, 1992, 4 (06) :863-879
[50]   A NEURAL NETWORK APPROACH TO STATISTICAL PATTERN-CLASSIFICATION BY SEMIPARAMETRIC ESTIMATION OF PROBABILITY DENSITY-FUNCTIONS [J].
TRAVEN, HGC .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1991, 2 (03) :366-377