ON THE CONVERGENCE PROPERTIES OF THE HOPFIELD MODEL

被引:79
作者
BRUCK, J
机构
[1] IBM Research Division, Almaden Research Center, San Jose, CA 95120-6099
关键词
D O I
10.1109/5.58341
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The main contribution is showing that the known convergence properties of the Hopfield model can be reduced to a very simple case, for which we have an elementary proof. The convergence properties of the Hopfield model are dependent on the structure of the interconnections matrix W and the method by which the nodes are updated. Three cases are known: (1) convergence to a stable state when operating in a serial mode with symmetric W, (2) convergence to a cycle of length at most 2 when operating in a fully parallel mode with symmetric W, and (3) convergence to a cycle of length 4 when operating in a fully parallel mode with antisymmetric W. We review the three known results and prove that the fully parallel mode of operation is a special case of the serial mode of operation, for which we present an elementary proof. The elementary proof (one which does not involve the concept of an energy function) follows from the relations between the model and cuts in the graph. We also prove that the three known cases are the only interesting ones by exhibiting exponential lower bounds on the length of the cycles in the other cases. © 1990, IEEE
引用
收藏
页码:1579 / 1585
页数:7
相关论文
共 10 条
[1]   A GENERALIZED CONVERGENCE THEOREM FOR NEURAL NETWORKS [J].
BRUCK, J ;
GOODMAN, JW .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1988, 34 (05) :1089-1092
[2]   HARMONIC-ANALYSIS OF POLYNOMIAL THRESHOLD FUNCTIONS [J].
BRUCK, J .
SIAM JOURNAL ON DISCRETE MATHEMATICS, 1990, 3 (02) :168-177
[3]   ANTISYMMETRICAL NEURAL NETWORKS [J].
GOLES, E .
DISCRETE APPLIED MATHEMATICS, 1986, 13 (01) :97-100
[4]   DECREASING ENERGY FUNCTIONS AS A TOOL FOR STUDYING THRESHOLD NETWORKS [J].
GOLESCHACC, E ;
FOGELMANSOULIE, F ;
PELLEGRIN, D .
DISCRETE APPLIED MATHEMATICS, 1985, 12 (03) :261-277
[5]  
Golomb S., 1982, SHIFT REGISTER SEQUE
[6]  
HOPFIELD JJ, 1985, BIOL CYBERN, V52, P141
[7]   NEURAL NETWORKS AND PHYSICAL SYSTEMS WITH EMERGENT COLLECTIVE COMPUTATIONAL ABILITIES [J].
HOPFIELD, JJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1982, 79 (08) :2554-2558
[8]  
Papadimitriou C. H., 1998, COMBINATORIAL OPTIMI
[9]  
PICARD JC, 1974, NETWORKS, V5, P357
[10]  
Rudeanu S., 1968, BOOLEAN METHODS OPER